Want to get more customers, make more profit & save 100s of hours with AI? https://go.juliangoldie.com/ai-profit-boardroom
Get a FREE AI Course + Community +1,000 AI Agents 👉 https://www.skool.com/ai-seo-with-julian-goldie-1553/about
🤖 Need AI Automation Services? Book a FREE AI Discovery Session Here: https://juliangoldieaiautomation.com/
🚀 Get a FREE SEO strategy Session + Discount Now: https://go.juliangoldie.com/strategy-session
🤯 Want more money, traffic and sales from SEO? Join the SEO Elite Circle👇
https://go.juliangoldie.com/register
Click below for FREE access to ✅ 50 FREE AI SEO TOOLS 🔥 200+ AI SEO Prompts! 📈 FREE AI SEO COMMUNITY with 2,000 SEOs ! 🚀 Free AI SEO Course 🏆 Plus TODAY's Video NOTES...
https://go.juliangoldie.com/chat-gpt-prompts
- Want a Custom GPT built? Order here: https://kwnyzkju.manus.space/
- Join our FREE AI SEO Accelerator here: https://www.facebook.com/groups/aiseomastermind
- Need consulting? Book a call with us here: https://link.juliangoldie.com/widget/bookings/seo-gameplanesov12
Оглавление (59 сегментов)
Segment 1 (00:00 - 05:00)
browser use AI agents course. Learn how to build and automate anything. Today, I'm going to show you five insane browser AI agents that can automate literally anything on your computer. These new AI agents can click, type, book flights, create presentations, and even post on social media for you. What I'm about to show you will completely change how you work online forever. And the best part, most of these are completely free and you can set them up in minutes. This isn't just another AI tool. This is like having a digital employee that never sleeps. Today, I'm going to show you exactly how to use browser use web UI, which is basically a powerful AI agent you can set up for free directly on your browser, and this can just go off and do whatever you tell it to do. Now, it's really easy to set up. I'll show you how to do that in a second. But what I'm going to show you today shocked me. Now, what you will see is that you can actually plug in the agent type and you can select custom, for example. You can actually give it multiple steps to complete the action. And also here you can see that you can select between using vision and not using vision which means that the AI can see your screen or not see your screen if you disable it. From here you can actually select multiple different APIs. And this is where DeepS version 3 comes into play because you can select DeepS as the API and then plug in your API key directly here. And you can also choose between using your own browser or using a browser from web UI. And additionally you can just give it tasks. So, I'm going to show you some of the best tasks that you can use to program this today, but it's pretty amazing what you can do and you can also record your screen along the way so you can see how it works and that sort of thing. So, we're going to get straight into it. The way that you set it up is by using these GitHub instructions. So, you're going to go to browser use web UI. You'll get all the video notes from today if you want to follow and steal this process step by step along with my best prompts. And directly inside here, you can see that you just need Python 3. 1 or higher installed. And then from here, you go to your terminal, copy each instruction as you can see, and then get it set up. Um, once you have that set up, you'll be able to host it locally like you can see right here. So, you can see I literally just follow the steps step by step. It wasn't actually working for me at one point. So, what I did was I actually just pasted in any problems that I had. So, you can do the same like you can do it inside Claude. Just paste any issues that you're finding when it comes to setting this up. And then once you've done, it will give you instructions on exactly how to set that up and correct any problems you have along the way. Pretty easy and simple to do. And now we can get round to getting this set up. So what we're going to do from here is we're going to select custom as the agent type. Then you can choose between how many maximum steps you want it to use to complete. So we're just going to stick with the default of 100. Now if you're using DeepS version 3 actually doesn't have vision, so you don't need to use that. And then from here we can say, okay, we'll use Deepseek API and we'll select the model name. Now, I'm also going to show you how to use this with a free API in a second, but we'll come on to that later. And if you want to get an API from Deepsee, you can just go to deepseat. com. The API is just extremely cheap. You can see that I actually topped the balance of $10 a week ago. This only cost me 2 cents so far to use it. So, the actual monthly expenses is ridiculously cheap. From here, we're going to create a new API key. Let's create the API key like so. We'll copy that. Plug it into browser web UI. Here we go. And then you're going to select the model name. If you want to use Deepseek version 3, then you would select Deepseek hyphen chat. So you just type that in inside the model name. And then from here, you've got the API details ready to go. Now, what you'll see inside the settings here is that you can actually decide whether you want to use your own browser or the browser from browser use web UI. I'll show you how that works in a second. You can also, for example, select the browser width and height. And here's what interesting is you can actually save the recording sessions. So, for example, when the AI agent is using your browser, you can record the whole screen and it'll be like a screencast recording right here. And then, for example, if we use a default task, go to google. com and type open AI, click search, and give me the first URL. We can run the agent like so. Let's test this bad boy out. I'm just going to go with the default prompt, which is go to google. com, type open AI, click search, and give me the first URL. Right? And then we can hit run agent like so. Now, you can see it's navigating to its browser. And now on google. com. It's analyzing the page using vision. It's typing in openai. com. Again, my hands are in the air. I'm not touching this. And you can see here that now it's grabbing the URL from the page. It's quite interesting how it actually analyzes the page. So you can see it selects different parts. And if we go to the last recording over here, we can actually play that back and we can see what happened. So this is the video recording. It's basically recorded a screencast, done its magic. I actually think this would be quite interesting for a YouTube video. Like you could create an AI voice to go over this YouTube video and then you could have the AI voice over talking about whatever tool you were talking about and actually using the screen at the same time. And then inside the recording section, if we scroll down, you can see that it gave us the final result. So it copied the main URL from Google, which is what we wanted it to do. It's got some information on
Segment 2 (05:00 - 10:00)
errors, model thoughts. You can see it thought process right there. Let's post that into a new chat. Let's have a look what happened here. So you can actually see the thought process here, what it did, how it worked, etc. And yeah, pretty amazing. The thing is as well because it's interacting with my own browser because I changed the browser settings to So let's try something else now. I'm going to say I'm going to change this to use own browser like you can see here. Then we're going to go to task settings and I'm going to say go to HFS then go to keyword explorer then do some keyword research for my SEO agency. We're going to hit run agent. Honestly, from what I've seen though, using your own browser doesn't work great. So, I'm going to disable that cuz it doesn't seem to work so well. Whereas, it seems to work much better in its own browser. And then from here, we're going to say, okay, go to Google and I'm going to try this, right? So, we'll see whether we can test the limits of this. I'm going to say go to google. com, type in best SEO agency Japan. Give me a content outline based on my competitors for SEO. And from here as well, I'm going to say you're an SEO expert, right? So, we're going to hit run agent. See how it performs. So, it's gone directly to Google. It's typed in the keyword. As you can see, now it has come up with a notification. I don't know whether it's going to be able to block that notification or not. It has managed to get around that. Now, there's another popup and you can see that it's looking through the page. So, it's typed in the keyword best SEO agency Japan. Pretty wild to see like the AI just navigating the internet and analyzing exactly what it needs to do, etc. Seems to be changing the language right now. I don't know what language he wants to speak. It's It looks like it's changing it to English. There we go. The thing that I've noticed with this is it can follow a lot more steps. So, for example, if you're using something like Claude Computer Agent, it seems to break within a couple of steps, right? It doesn't seem to get very far. Whereas with browser use web UI, it seems to be a lot more competent at figuring out, okay, here's what to do. Here's where to go. And it can just perform more complex actions. Now, it's actually created the content outline here. So let's have a look see what we get. Interestingly it's written it down in markdown which is not necessary but let's have a look here what we got. It says I've analyzed the search results and I've identified some of the top SEO agencies in Japan. Here's a content outline based on my analysis introduction list of the top SEO agencies. So it's actually taken like the names of the agencies on the list. Then it said provide a brief overview of each agency highlighting their strengths and services key SEO services. How to choose the right SEO agency. pretty interesting that it has actually done the research, figured it all out for us and then based on that it started creating the content analysis. Now I'm going to say okay go to google. com find me the cheapest flights from Bangkok to UK and I'm going to change this now to you're an executive assistant. One of the other benefits of this as well is because you're actually using the latest APIs, whether that's Deep Seek version 3, whether that's for example Gemini Flash Experimental, because you're using these newer APIs, you can get much better outputs, right? Whereas, for example, if you were using something like Claw Computer Use, that's not been updated for months since it released and so you're not going to get the same level of outputs. I'm really determined to try and use my own browser, but I just don't think it's going to work. Let's click this as well. See what that does. Yeah, it just has errors. So when you use a browser, it doesn't seem to work. Maybe I need to allow something inside the settings there, but it just doesn't work so far. So switch off those two features and then we'll get back to it and let's see what it does now. How good is it at actual research tasks? Because for example, you can imagine like these sort of basic admin tasks where someone just needs to click buttons and it's not a very highly skilled job can now be replaced by AI, right? And the thing is these are the worst that these AI agents are ever going to be. And you can do this for free as well. These AI agents are only going to improve over this year. I can imagine ChatGpt, Claude, Google, they're all going to be releasing their own agents pretty soon, maybe within the next couple of months or so. But then number two, if you've got the choice between getting an AI agent to do it for you or hiring a virtual assistant that might cost like $5, $10, $15 an hour, of course, you're going to go directly with the AI agent, right? just makes sense especially for businesses or people using executive assistance. So it's really important to figure this out because you got to understand okay where's the industry going but then also okay where does what jobs are actually still going to be necessary for example if I look over the last sort of past 12 months 2 years or so since using chat GBT we don't really hire human writers anymore we don't hire photographers anymore all the editing on my main AI channel is done with AI using dscript not with a human right and we're seeing it unroll in real time right so it's really important to experiment with these even if they're not 100% % perfect simply because it will change the world and things are only going one way if you can see what's happening right now in real time. For example, like even for coding, I wouldn't hire a developer anymore, especially a front- end one because I can develop whatever tool or
Segment 3 (10:00 - 15:00)
whatever SAS tool I need in the space of minutes using something like BoltDIY and that won't cost me a penny and also it's a lot faster. It's super important to figure this stuff out now. So, let's have a look at this. So, we've got the output from the browser. They said, "The cheapest flight I found from Bangkok to the UK is to London for 23,000. Departing on May the 12th, returning on May 25th with one stop and a total travel time of 19 hours 50 minutes. The flights are with Shanghai Airlines and China Eastern. " Right. And that's a pretty detailed and comprehensive analysis. We could have given it the dates and how much we wanted to spend, but honestly, is that cheaper, faster, and easier than using an assistant to go and do it for you? Absolutely. And so, give it 6 months, give it 12 months. Who knows what the world's going to look like. So, thanks so much for watching. If you want to get full access to all the SAPs and instructions on exactly how to set this up, you can directly inside my free SEO course link in the comments description. What you also find inside the video notes from today is 50 use cases and prompts of exactly how you can use this in real time. So, it's very practical and that's included inside my free SEO course. This comes with over lessons on how to do SEO with step-by-step tutorials, SAPs, tips, templates, 53 SEO tools and a community of 3,000 SEOs that there to help you with any questions you have. Now, if you scroll down to the free SEO tools section, you'll find loads of cool stuff, including the video notes from today, which are directly down here, and you can get access to all the links, all the tips, all the prompts, and the SAP from today step by step directly inside the comments description of this video. Additionally, if you want to get a free SEO strategy session that shows you how we take websites from zero to 145,000 bits a month and generate hundreds of thousands of dollars in sales on autopilot, feel free to book that in. We'll show you on this free link building acceleration session. You'll get a free SEO domination plan the secrets SEO link building. We answer any questions you have onetoone. You'll learn the best link building strategy for your website, plus how to quickly outrank your competitors link building and how to 10x the SEO traffic based on what's working for us. Feel free to get that. Appreciate you watching as always. See you on the next one. Bye. Today I'm going to show you a new browser use AI agent to build and automate basically anything. And this is available with a free tool called R code inside Visual Studio Code. Right now the way this works for example let me show you some examples of this morning and how I've used it. So for example here we said go to Claude and have some cheeky SEO banter and he went over started typing into Claude directly. You can also see some other examples. So, if we go over here and we go to here, you can see we actually got it to start tweeting for us, right? So, we said, "Go to Twitter and post a tweet. " Write a short inspiring quote about AI and post a tweet. No hashtags. And then it just went off and did it. Look at that. Drafted the tweet. Ready to go. All right. Now, if you want to learn how to do this, I've got the SAP inside the AR profit boardroom link in the comments description. Essentially, what you want to do is you want to get Visual Studio Code and then you're going to go to Root Code, right? So you install root code inside the extension section. Then you're going to go to root code and inside here you're going to select anthropic. So you're going to be using anthropic which is Claude's AI to build this. And you see how it supports images, computer use, and prompt caching. Right? You want to make sure that you have auto approved setup. And there are two ways to set this up. Basically, you can have the AI inside a completely separate Chrome browser where nothing is logged in or you can actually have it log into a Chrome browser where it has all your login directly, right? So, you can go into Twitter or whatever. If you want to set this up and you want to know, okay, which prompts should I use for which, then you can get that inside the AR profit boardroom SAP. And then if we scroll down here, this is the most important part. We're using the enable browser tool section, right? And this is how we're going to link the computer use, right? You see how it says browser and computer user and then enable browser tool. We're going to use this screenshot quality I would keep quite low. I'm just keeping it high cuz I'm doing a video. Obviously, it uses more tokens. So, that's the price I'm going to pay for you. But you can see here we've got the setup and then the local host. You get this set up with the SP from the AR profit boardroom. We'll hit save. We'll hit done. Boom. Shaka laka. Let's get off to the races. All right. So, for example, if we say now using this computer use agent, right? Now, you can just brow it can go. or we can browse to use whatever. So, we're going to go to julianoldie. com and do a comparator analysis of Julian Goldie. Make sure that you have Claude 3. 7 Sonic, which is the latest API key set up from Claude. Then, we're going to get it to do this action right here. And it says, I need to perform competitor analysis. Here's what I'm going to do. Blah blah. And it's going to run into this separate browser, right? So, this is a browser where it's running in debug mode. It is logged in as me. So, you can see I'm logged into WordPress on this website. And it's using this to really scan through the website, scroll through, etc. Right? So you can see a log of what it's doing right here. So I can see Julian Gold is website. Here's what I need to do. And it'll figure out number one when it needs to use the browse action. So it's not going to use it every time, but only when it has certain tasks that require that. Smart enough to figure that out. And then I need to check out the services page, right? So it goes to the about me page
Segment 4 (15:00 - 20:00)
goes to the services, etc. And it just starts scrolling through my website, right? It can see the page. It can interact with the page. And it can also scrape websites. It can Google stuff. It can go on to YouTube. Like, it's pretty wild what it can do. In fact, I don't really see a limit to it. Now, I'll show you some other use cases in a minute, but let's wait for this task to load. So, now it's right browse use action. Here's what we're going to do. Blah blah. It will just start going off and doing its magic. It's scary to watch this stuff because it's so powerful and it's too good at what it does, right? So, if we open up now, it's scanning through my quesy gander here, seeing what's popping. The cool thing about this as well is like you don't need to be able to code. You don't need any technical expertise to get this set up. Super easy and simple and it really doesn't take a long time to figure it all out. Like for me, for example, I have no real coding experience, but even I can do this, right? So, it's gathering all the information here. It's having a look through the AI automation section, the case studies. It's browsing the user. It's let me check the Elite Circle page as well to see what premium services Julian offers. And it can scrape the web. It can, for example, read your pages. It can do a full background check on any of your competitors. And it is unbelievably easy to use. So, it's doing all of this whilst I'm talking. Like, I'm not clicking anything. I'm not doing anything. The other thing I like about this is like I can do stuff in a separate Chrome browser. So whilst it's doing that, if I open up this browser and I just go over to Twitter or whatever, I can do that pretty easily and I don't need to worry about like this interacting with the other Chrome browser. So it's still interacting here, but we've got our separate Chrome browser that we can do work in as well, right? So you can work side by side with your AI agents whilst you're doing this stuff. All right. So you can also stop the task anytime you want and you can say okay summarize the data that you got. So if you ever want to stop it or take control you can as well. You can also take control of the instance. So for example I could start navigating through this as well and direct the AI how I want to do things. But you can see here how it's doing a full competitor analysis based on the information it's scraped. Right? This is based on explanation of Julian Goldie. He's a competitor analysis. Here's a business overview, core services and link building. Here's how much he charges per link. So, it actually knows what we charge, how much we charge, his team, how big his team is, client testimonials, etc. Marketing approach, right? So, it's even broken down my funnels there and exactly how we get clients. And it's done all of that really quickly and powerful. You don't need any fancy graphics card or anything like that. I'm just running this on a M3 Pro, right? But I don't think because you're using an API key because this is not really running you're not running the AI locally, right? You're just running the coder locally. I don't think there's going to be many issues if you've got like a slightly older computer. And then we've got the full competitor analysis broken down right there based on the images and the content is scraped. Right. So let's do another one right now and we'll open up a new tab. So I'm just going to click on the plus sign at the top here to open up a new task. And then we say do a Google search for SEO training Japan. and based on that do an SEO optimized content outline. Right? So you can create your own like sort of SEO outline tool and reverse engineer your competitors on Google by using the AI. And also if you click this AI in the bottom right here, it will actually improve the prompt, right? So it will create something much more comprehensive. So if we hit enter now, we've got a much more comprehensive prompt to start understanding and creating the content outline from Google. It's going to start searching Google now. So you can see here for example it's typed in SEO training Japan. It's actually found my website ranking number one for that keyword and it should just start opening up different pages from the SERs analyzing their content and then reverse engineering that content to figure out okay what should we be talking about here as well. Just going to close that. So you can see how it's opening up my competitors is analyzing their content trying to understand it etc. looking through the Serbs and using that information to really understand our competitors and create the best outline. Now, this is pretty cheap as well. This cost 6 cents to do that and it's gathering more information to figure all this out and it is really cheap. The other cool thing about this as well is it's quite similar to cursor and other tools like that, but you don't have to pay for this, right? All you're paying for is the API key and you can get a free API key as well with Google and open router as well. So now it's opening up like the people also ask section as well. So you can see it's ask it's going through there looking through the people also ask questions gathering some relevant questions for the content outline and then it's let me try some more searches here. Now if you ever want to stop this at certain points right let
Segment 5 (20:00 - 25:00)
me show you how to do that. So wait for this to load. We can click cancel there. we can override it and we can say okay now based on what create the SEO content outline and then because it's done the research it's looked through a couple of our top ranking competitors it can use that to create an SEO outline so you see here how it's creating that file and it's called the SEO training Japan outline MD that's actually a text file like you can see right here with the outputs from the agent and it's actually planned out exactly how to create the content, what to write about, some topics, etc. And it's doing that all based on our competitors that is scraped from Google. So, it's pretty crazy what it can do and how powerful it is. You might be asking like, okay, can this, for example, all right, we've got the output here, but you might be asking, for example, okay, can this create videos and stuff for you? So, that's a really interesting point. So you could actually do a screen share where you just ask the AI to navigate around whatever tool you're talking about on YouTube and then screen share that. The other option that you have is you could actually use this process to then operate something like dcript and a custom GPC. I've got videos on YouTube about that and also inside the AI profit boardroom. But you can basically ask the AI to go off use chat GPT create a video and then using that plug it into dcript and export it. You could get it to do it that way. I'm not going to do that on this video, but you get the idea. And then we've got the setup right here for the SEO outline. Pretty simple and easy to get started. It's recommended the sections, the content, who the audience is, and then also how to set this up. All right, so you got a clear hierarchal outline inside the MD file. Pretty simple and easy. All right, if we click on start new task now, we can start doing other stuff. Some people saying this sort of stuff it feels more powerful than manus. Yeah, if you check out my video on brew code versus manus, you can see the key differences. Just search YouTube for that if you want to find the information. But yeah, in many ways it is more powerful than manus cuz you don't have any daily limits as well and it's coding locally and you can switch between whatever API key you want. So now we're going to say all right, do a competitor analysis of Julian Goldie's YouTube channel. Let's see if we can do that. So, it's going to open up YouTube, but it's gone to the wrong page here. So, now it's going to try a different way. So, it's actually going to search YouTube for Julian Goldie. It's opening up a new tab. Search YouTube for Julian Goldie. And it should just start going through YouTube and start having a look through what's going on here. Yeah, you can see here. It's great. I found Julian's YouTube channel. Here's the information about him. Let's start researching all this shorts, etc. And it's given us a breakdown of the latest video titles plus the views, right? So, it can scrape YouTube, it can scrape Google. It's actually found the most popular videos as well. So, for example, like this one on Deep Seek that got 2K views and it's basically like a little research assistant that can interact with the web light, right? But the difference here as well is unlike a human, it's not going to get stuck on the weekends or bogged down in questions. It's just going to figure stuff out. It's very proactive as well. And then it's even looking for some of our competitors as well, so you can see. So, it said top SEO AI YouTube channel competitors to Julian Goldie. And then it's found a bunch of options right there. So, really fast, really powerful, and incredibly easy to get this set up and running. And now it's going through some different articles, having a scan through what we're doing, etc. Again, you can cancel this at any time as well. All right. So, if you don't want it to keep running or you don't want to rack up the tokens or you just you're in a rush and you're trying to get this done like you can see on a YouTube video, then we can say, okay, knowing what you know, and this time around I'm just going to create like a nice little onepage. Create a onepage beautiful presentation in HTML that explains everything you know about Julian Goldie based on your research. All right, include real facts, figures, plus make it super modern. All right, so basically what it's done is it's scraped the information from the web. And then what we're going to do from here is we're going to code out a page locally on our laptop based on that information and the number of views and everything else it's done right here plus all the titles. And now you can see it's beginning to code out the HTML. All right, so it's connected to the internet. It can interact with log into accounts and do different stuff. It can do really amazing research tasks. It can read pages. It can interact with pages and it can code locally. Plus, once it's actually finished coding out this HTML file, what it's going to do is it's going to open that up in a browser as well so that we can preview it and
Segment 6 (25:00 - 30:00)
take a look at the content. Now, some people might be asking, okay, can you run this with a different API key? So, for example, could you set up like a computer use local API and then use that inside here? So if we go inside the API provider, we can actually select OAM and O Lama allows you to run APIs locally and then you could configure this so that it's using your AI from O Lama, right? So it's broken that down right here as you can see. So it's figured out, okay, this is how many subscribers Julian Goldie has. Here's a competitor analysis. Here's the experience. Here's what he focuses on. Here's his top video and how many views it got, plus a title, plus an overview of what he does. the top content that he talks about. And if we go back now to Visual Studio Code, it's done its magic. All right, it's even showing us a preview of what the page looks like. Scrolling through it, etc. And that whole competitor analysis, well, this is a really interesting thing is it cost us 40, right? But if you paid for a really in-depth competitor analysis with a nice presentation from a virtual assistant, it's going to cost you maybe minimum $10. But if you want it done properly, maybe like a h 100red or $200 or $500. And so the arbitrage between using an AI to do these tasks versus getting a human to do it is massive, right? Like you could sell this as a service. Like you could list a fiber gig and say, "Right, I'm going to do competitive research and I'll create you a beautifully designed presentation based on that, right? And then you just go off and get something like Root Code to do it on autopilot whilst you watch Netflix, whatever you want, right? Like you could sell all the stuff that you do inside R code with this computer use agent. You could easily sell as a service. I don't think that would be an issue at all. We could also do other stuff as well. So if we open up a new tab here and we'll say okay now use HFS to find low competition keywords for a dentist. The other cool thing about this is you don't need to train the AI, right? So if you were, for example, had a human working, you'd have to quality control the work, check they're doing it properly, check in with them, manage them, give them emotional support as well, which is like genuinely that is quite a draining thing as a manager. And then you'd have to train it, you'd have to set up good SPs, etc. Whereas with this sort of AI, you can say, just go off and do this, mate. I don't want to train you. coach you on the weekends. I don't want to quality control you and everything else. and it'll actually just go off and do it. Like you can see, right? And it will go off and do it in the background. Like you don't even have to be there in his presence to know that the job is being done. So it's pretty scary what this can do cuz it's so powerful and it's so far advanced in such a short space of time. Bear in mind, I think about a year ago computer use came out and it wasn't that good. It might have even been longer ago than that. Whereas now, this is well and truly getting towards something that's better than a human. If I asked the average Joe to just go off and do some AHF's keyword research, he wouldn't know what I'm talking about. Whereas the AI knew exactly what we're talking about, what we need to do, and how we need to do it, it just gets on with the job, right? No messing around. No taking Sundays off. And you can see now it's come back to us with the keywords. So, it's given us a list of relevant keywords right here. And it's searching for lower keyword difficulty stuff, right? So, it's actually going to search for KD. And then it's found a bunch of KD0 keywords right here for SEO, right? And these are all like genuine keywords we could actually use. So you get the point. Like it's super powerful. It's unbelievably easy to use. It works autonomously. It doesn't require training and it's pretty cheap compared to using an actual real person. The other thing as well, like for example, if you learn a keyword research report like this for someone, we've charged in the past like $500, $250 for keyword research reports, but now you can automate them for the space of 20 like you can see directly inside Rukode and it will probably do as good if not better a job than most SEO agencies, right? The other thing to note here is and this is another scary thing especially for employees is an AI agent. It doesn't require a employment contract and it also doesn't require any sort of rights. It doesn't take days off. It doesn't get sick. It doesn't get stopped. You could open up like five different windows and have different AI agents doing all sorts of different stuff and it could just keep going, right? whereas obviously a human never could do that and you would never expect a human to be able to do that but you get the point. And then you've got a really nicely a really good report right here. This written down in markup. So if we go back to we're going to go over to markdown lipreview. com. We'll paste in
Segment 7 (30:00 - 35:00)
that and we've got the report right there ready to go. So this is the research report from the AI agent and we asked it to find low competition keywords for dentists. She did that. For example, we found a bunch of KD zeros right there. Very low difficulty, medium difficulty, and analysis and recommendations. Right? This has made it super actionable as well. It's like dentist near me. It's a little bit competitive, but here's what you want to go for. It's giving us some longtail stuff and also some content strategy recommendations. It's a pretty complex task as well. If you were trying to do this yourself, it's probably going to take about 30 minutes to 1 hour, whereas now we can just get it done instantly. The same for example with coding, right? Coding would take you years. It's quite a complex task, but you can mock up like a SAS tool in the space of 30 minutes or 60 minutes inside R code and you don't need to learn code or anything like that. So, it's not that simple. Yeah. One final one that I'll show you for social media management. If we say, okay, go to Twitter and draft a tweet that's a short inspiring quote about AI. No need to post it. All right. It's going to figure out what to do and then it's just going to go off and do it. So, let's see what it does next. see if it figures it out. So, it's posting for my AICO Mastery account, which is a totally separate account that I just use for test purposes. And it's typed in the quote, as you can see. I said, "The most powerful AI isn't the one that thinks like a human, but the one that helps humans think better. " Pretty powerful quote, to be fair. If we hit post like boom shakalaka, we've got it done. You can see the tweet going live right there. That is a nice little tweet. probably better than most of the trash out there. So, you get the point. Like, it can do really complex tasks. It can build SAS tools, games, apps, websites. It can tweet for you. It can use SAS tools for you directly. It can automate stuff. It doesn't take weekends off. And it's really easy to use. You just like, go off and do that. Boom. Let's And it cost you seven cents to do that. What would a social media agency charge you? Thousands of dollars per month. and you can just post all day for 7 cents per tweet. It's ridiculous. So, thanks so much for watching. If you want to get the SP for setting all this stuff up, you can get that inside the AI profit boardroom link in the comments description. This is an amazing community that comes with 508 members, all dedicated to making more money and saving time with AI. Like you can see, prices is going up at the end of the month. So, make sure you sign up now before prices go up. Inside here, you get all my best automations for email and content automation, social media and video, AI agents and workflows, AO automations, NA10 templates and workflows, and all my best SAPs, including the SAP for setting up R code, as you can see right here. Additionally, inside the community, you can post any questions you have, and we're constantly updating this. Like you can see right here, all these updates went just over the last few days, and we're pretty much updating this every single day. And also, we run weekly Q& A calls. You can jump on the Q& A calls, ask us any questions you have, and if you can't make the Q& A calls, it's okay. You can just watch back the Q& A call recording is right here. Appreciate it. Make sure you get that before prices go up. And additionally, if you want to get a free onetoone SEO strategy session that shows you how we take websites from zero to 145,000 bits this month, feel free to get that link in the comments in the description. We'll show you how we take websites from zero to this. and also generate hundreds of thousands of dollars in sales. And you'll basically get a custom tailored SEO game plan so you can generate more leads, sales, and profits from your website. Plus, you'll be able to ask any questions you have, and you'll learn the best SEO strategies to 10x SEO traffic based on what's working for us and our happy clients like you can see right here. Feel free to get that link in the comments and the description. This new claw browser use AI agent can help you build and automate anything. For example, if I say open up my browser and search the web for the latest AI news, what you can see on the left hand side here is it's navigating to google. com with the browser navigate MCP, which is completely free. And you can basically automate anything that you would do on the web using this free MCP. And in the next few minutes, you'll learn how to set up an AI that can browse the web, search for information, post content automatically. You'll learn the secret way to connect Claude to both your browser and your computer files simultaneously. and you'll discover exactly how to install the extension and configure the MCP server with zero technical knowledge. And yes, this is all working whilst I'm talking directly to you. Imagine having an AI assistant that can navigate websites, build tools, and access your local files all for free. Watch now to see how I built mine in just a few clicks. Today, I'm going to show you how to basically automate anything with this brand new browser use agent. So, for example, here we've got Claude desktop open. And if we say, okay, search my browser for the prices of the S& P 500, right? And essentially what we've got here is a MCP agent that can open
Segment 8 (35:00 - 40:00)
our browser and then start using it. So as an example, here's an example. So we said, open my browser and check the price of the S& P 500, right? And you see here it used browser navigate to have a look and search for that using our SMP. So let me show you an example. There we go. It's using browser navigate MCP. And then if we check in the background over here, you can see that it's actually navigated to browser use agent. So it can use the browser in the background to navigate and search stuff. So, if we go back inside here and we say, "Okay, open my browser and check the latest headlines for AI news, right? " And so, we've got Claude Desktop over here. And Claude Desktop is going to browse the web inside Claw Desktop and then give us the results, right? And I'll show you exactly how to set this up for free in a second. So, it's now browsing the web. It's looking for AI news. You can see an example right here. It's pulling in those headlines and boom, we're good to go. Now it's browsing the web, navigating through Reuters. com. And again, this AI agent is not just easy to use, it's also super fast and it's completely free to use. So, it's absolutely crazy how powerful it is and how easy it is to set up. I'll show you how to set up in a second, but you can see essentially here's how it's working, what it's doing, etc. It's navigating our browser, looking at our browser in the background, and searching for whatever we want. So, I'm just going to stop that for now. But you see how easy this is to set up. And now I'm going to run through exactly how to install it. So, if you want to install this bad boy, basically what you're going to do is you're going to go over to, let me show you here. We've got the notes inside the AI success lab. So, you're going to go to docs. browsermcp. io. And this is basically a free MCP server that you can set up to browse your web. And it's using claw desktop to do it, right? So really easy and simple. And then you can click on install in the top right. And basically what you're going to get here is the MCP server configuration. Now when you're setting this up, essentially you would just grab this code. So you can just copy this. I'll put it inside the AI success lab. link in the comments description if you want to get this stuff just for free and you just want to steal it from me and copy and paste this stuff. There you go. So, you can just copy and paste this if you want to. And essentially, you're going to go into Claude, make sure you've got developer settings set on. So, allow developer settings. Then, you're going to click on developer, open MCP log file, and you can see the details of the MCPS you're using. Right now, inside Claude desktop, what you want to do is go over to Claude in the navigation menu. Then you're going to click on settings. And then inside your settings, you're going to go to developer. And you can see the MCPS I've got installed. Now, if you want to install MCPS on claw desktop, pretty simple. So, you edit the config file like this. You can just open it with text edit and then you can install whatever you want. So, inside my settings, I have browser MCP set up. I also have Apple script set up as well. And I'll come on to exactly how these can work side by side because it's pretty crazy. And then we have the basic settings for MCP servers. Now, if you want to just steal that configuration from me, and again, these are all free MCPS. You don't have to pay for them. Then I will include that inside the AI success lab as well. So, you can just steal my full MCP setup if you want to. All right. Um, what this can essentially allow you to do is not just navigate the web, but also navigate your actual computer as well. So, let me show you exactly what we're doing here. Before we get on to that, what I also want to show you is inside your settings once you've set up the MCP config file. So, you're just going to hit save like that. You've got your MCP settings inside here. You're going to close that and then you would restart Claude and that's when it's going to trigger the MCP settings. So, pretty easy and simple to set up. Once you've done this, we've now got two different MCP settings set up, right? So, we've got Apple script and this can actually go straight inside your browser and start searching stuff. And then we also have, sorry, Apple script can connect to your laptop. So, it can work internally with your laptop and everything that's on there. And then we also have browsy's agent which can then surf the web for us, right? So we can do internal and external if that makes sense. And now you've allowed Claude to be like 10 times more powerful because it can interact with your laptop. browser etc. Okay. So for example here if we say okay what's my battery percentage? It's going to use the Apple script MCP as you can see here. And you see here it
Segment 9 (40:00 - 45:00)
says your battery is currently 96%. And we can see that is indeed 96% in the top right there. Right? So it's really easy to set this stuff up. Pretty crazy what you can do with it. And it really doesn't take a long time, right? All you have to do is allow the MCPS. Now, if you want a full setup on MCPS and how to build them, etc., I've also got another training inside the AI success lab. Link in the comments description if you just want to steal this stuff from me. But essentially between the Apple script MCP and the browser use MCP you are absolutely flying my friends and it's incredibly easy to do. So let's keep going. Now what we're going to do from here is we're going to go back into Claude and we'll pull up example of what we can do here. So I'm now going to say okay open my browser post a cheeky tweet on Twitter. All right. And I'm just going to post this on a test Twitter account, not my main account, right? About and we'll put about AI, something like that. So let's see if it can do that. So using browser, navigate to the twitter. com page. And you can see it is doing that over here. And then we just have to allow this MCP to use an external integration. So it says browser MCP review each action carefully before approving. Law cannot guarantee blah blah. I'm going to click allow always. But obviously you use this at your own risk, right? And now it's opening up Twitter and it's navigating as you can see right here. So this is the browse the mouse that it's using to navigate around. It's going to compose a tweet and what it's doing is actually if we click on the browser navigate tool here, it's actually like taking screenshots and just figuring out what's going on. It's like right great. Now I can see the tweet compose dialog is open. Let me type a cheeky tweet about AI. It's also got a browser wait function. So that allows it to wait before it starts posting. Let's try something else. So we're going to say, okay, go to bolt new and give me a prompt to build an SEO tool. So as you can see, it's now navigating to bolt new and then we've got the prompt that we can just check out, right? So it navigated to bolt. n new gave us the prompt plugged it and then we plug it in here to build out whatever tool we want. Right? So you can use this AI agent to instruct other AI agents to navigate to them and side by side you can do some pretty powerful stuff. Now also with the Apple script example what we could even do for example is let's wait for that to load. So it's now coding out the tool. You see the code over here. It's doing the data visualization. And again, I haven't paid to use this tool at all when you're setting up the extension as well. So, you do need the Chrome extension as you can see right here. And once you're setting up, you just need to connect, right? So, for example, if we go to my extensions, click on browser KPI, we can say connect from this tab or focus on one connected tab, etc. So, that tool is now building out and it's pretty simple and easy. What we could also do is we could go back to bolt. We can take an example app that we've already built out. What we can do is we could say to Apple script, take a screenshot of my browser and then give me some ideas on how I can improve this app built in Spanish C browser. So allow that to go. It's got a tool called browser screenshot. So, it's taken the screenshot and then it gives us some ideas on how to improve it. Switched to Spanish mode though for some reason, which is not ideal. I'm just going to be transparent with you there. I was going to say right in English, but it's basically taken a screenshot from the app that we've built as you can see inside bolt. And then it's given us some ideas on how we can improve that on the left hand side over here. Right? So it can read your screen, it can navigate, it can browse. It's not that great at interacting with social media platforms to be honest with you, but you get an idea and a feel of like how powerful this can be. And also it can interact like directly inside your laptop as well, right? So if we go inside here, we're using the Apple script MCP. We can say, okay, give me the last three files in my downloads from the last 24 hours. Something like that. And then I'm going to check your downloads folder for the most recent files from the last 24 hours using Apple script. It's setting up all the filters right here inside Finder as you can see. And also there's a bunch of different prompts you can actually use with Apple script. So if we go back inside a browser here, we'll go over to the claude MCP section inside the ASS lab. Then from here we're just going to for example create one of these. So for
Segment 10 (45:00 - 50:00)
example, show me all my files in my downloads folder for the past week. And you can see that's retrieved the information, right? So based on the information I've retrieved, here are the most recent files downloaded in the last 24 hours. And then it's found the examples of Flow Whisper that actually downloaded recently. Then we've also got two thumbnails for the live stream that I'm doing right now. And so like it can interact with your browser, with your local folders, and you're basically making an AI super agent that can interact with pretty much anything to get really good results, right? We got a question from Dennis Hugh, the legend, who says, "Can it make a list of all the podcasts I've been on, including yours? " Let's have a look. I'm pretty sure it could do something like that quite easily. So, let's try this out. We're going to say Ryan, open up my browser, find podcasts that then you it's been on recent. Something like that. Let's see if it can work its magic on this. So, it's going to open up the browser. It's using browser navigate and it's searching on Google to do that. So, if we go over here, you can see it's searching. And if we pull this to the right hand side, you can see it's navigating and checking the recent podcast appearances. All right. And then it's also searching Dennis U podcast guest 2025, etc. You can see on the right hand side, this is Dennis U and his podcast. And on the left hand side, here's a list of his recent podcast. So, let's pull this up and see if it's relevant. So it said recent Dennis U podcast appearances 2025. Shout out to Dennis, the Ask My Advisor Show, The Local Marketing Secrets, Speak with People, Neil Schaffer's podcast, etc. Even gives us the length as well. So let's just check that now. I'm just going to copy that and then plug it into Chrome. See if we can find that. There you have the podcast that I found. Right. So this is the Dennis Hugh podcast that we were talking about with Neil Schaffer 10 months ago. So fairly recent and that is definitely with Dennis U as you can see. So any basic stuff like that, anything that an executive assistant for example would do, this can now do and it's automated. It's really fast to do. So super powerful. I think if you wanted a really comprehensive list, yeah, you could use deep research or something like that. I think there's a research option here. So you can just go inside Claude and research like that. So yeah, essentially that's how you can use the MCP to do research, to post tweets for you, to check bolt new, to check your apps and rebuild them, etc. Pretty much anything that you can think of, you can now do with these MCPs and the interact both locally and also directly on the internet as well. It's crazy. They're liberated and autonomous. So, thanks so much for watching. If you want to get access to all the video notes from today, feel free to get that inside the AI success lab. link in the comments in description. Completely free. You can get all the notes from here. You can get the code that I've used to set all of this up. It is free to use inside Claude. And then additionally, if you want to upgrade and get, for example, an amazing community focused on making more money and saving time with AI, feel free to get that link in the comments description to the AI profit boardroom. Basically, if you want coaching, if you want support, if you want help implementing all this stuff, then feel free to check this out. this an amazing supportive community of great people who are focused on growing their business and scaling with AI. So, link in the comments description to the AI profit boardroom as well. And if you want to just get the free video notes plus links to everything, feel free to get that inside the AI success lab. Claude 4 has just dropped. I'm going to show you a way to basically turn the new version of Claude, which is the most powerful model in terms of coding the world has ever seen. If you're using Claude Opus, and you can turn it into an AI super agent. This is Claude Sonic 4 and you can get free access to it as an example of that. Let me show you. If we pull up the browser MCP, we'll connect right here. And then now we're going to say inside Claude Sonet 4. So make sure you have Claude Sonet 4 selected on the desktop of Claude. We'll say find me the cheapest flights from Manchester to Singapore with browser MCP. Hit enter on that. And now you can see it's navigating with the browser to find and research flights, right? So you can see it's using the MCP over here and the browser use agent to navigate to do the snapshots etc and start searching right so it's even moving the cursor typing stuff in and with the new version of Claude Sonic 4 this is actually more powerful than ever before right and it's quite easy to get it to actually work is far better than it ever used to be because of this new update right now we're also getting step-by-step instructions for this can work in the background and then you can go off and do stuff whilst and this free MCP super agent is just going off and doing its magic. Now, I'm going to show you exactly how to set this up for free as well. So, you can access it for free. It won't cost you anything. Let me show you how in a second, but it's super
Segment 11 (50:00 - 55:00)
powerful, right? So, for example, here what you want to do to set this up is you're going to download Clawude Desktop for free, right? all the links inside the AI success lab if you want to learn how to do this stuff for free. But essentially, you're just going to download Clawude Desktop for free. You can download it right here. It's free to access. You can even see, for example, that you can download it for Mac, Windows, etc. And this works on both Mac and Windows. Then from here, what you're going to do is create the MCP config file and download this browser MCP extension for Google Chrome. That's going to allow the MCP to connect and it's going to allow you to control your browser with claw directly inside desktop. Now, you might be wondering, okay, at this point, is it free to use? Can you use this stuff for free? And as you can see right here, you can access it for free right now. If you want to set up the MCP server, all you do is you go inside Claude here and then inside Claude, you're going to go to file and select the Claude section here, go to settings. Inside your settings, go to developer and then edit the config. Right now, inside this config, you just open with text edit like you can see and you just paste in the MCP details of anything that you want to use. So, for example, if we go back to the AI success lab where you can get the instructions, you can just copy and paste this stuff if you want to enable it, right? That's all you paste into the MCP. So, you can steal it from me completely for free. And then basically that's going to create the MCP config file. Then you can configure it and you just need to restart Claude once you've done, right? Pretty simple and easy. That's basically how you can set up this AI super agent. And then once you've reloaded it, you're going to see your MCPs enabled. Like you can see these are all running. And now we can connect Claude to our desktop, to our browser, but also even to our laptop. Right. So if we pull this up as an example, let me show you. It's found the flights from Manchester to Singapore for May the 24th, as you can see. So it did all the research right there. And then if we want to pull some examples of, you know, what we can do with this, I'm just going to enable the browser use MCP, we'll connect. So you've got to make sure you connect it. And then if you want to see some examples of what we can do, let's pull this up. So side by side, you got claw desktop on the left hand side and you got the browser on the right hand side. So if we start a new chat here, then we're going to say, write a blog for me in Markdown via Markdown live preview using browser MCP. We'll hit enter. It's navigated to Markdown live preview. He's editing the text box and going off and doing his magic. Carlos says, "I just signed off the AI profit boardroom finally. Welcome on board, mate. Great to see you here, legend. " Let's try another example. And so if we go inside here, we'll say, "Okay, go to LinkedIn. Find me some CEOs who run SEO agencies. " Now it's navigating to LinkedIn. And you can see whether it's using the MCP or not by this B sign. Right? So it says browser, navigate, request, response. Now it's taking screenshots. So you can actually take screenshots of the screen. Click through, etc., find the search buttons. With Claude Sonic 4, it just seems a lot faster and seamless than it's ever been before. So now it's perfect. I can see the search results showing CEOs of SEO agencies. Let me scroll down to see more results and get a comprehensive list. And it's actually broke down all the list of CEOs it's found from LinkedIn directly. Right. So if we go on the left hand side here, it's got the names of the CEOs that it's found right here. So for example, Waja Ali, shout out to Waja. He's over here. I'm sure I've butchered his name. Sorry about that, mate. But you can see here that it's collected all the details of the leads that it found on LinkedIn, screenshotted it, and then come back to us with the details. Right? So it's pretty simple and easy. it can operate your browser for you. And you can see an example of the screenshot here. So you can see how it's operating the screenshots, what it sees on the screen, and that this isn't just like BSing you. It's actually finding information and doing stuff for you, which is pretty crazy. Now, it does have some limitations, right? So, for example, I'm pretty sure if we say, okay, post an interesting tweet, it's going to struggle with that. So, if we go over to Twitter here, for some reason, it really struggles with Twitter. So, I'll switch to a test account and then if we go inside here and just I want to show you its limitations because it's not perfect. So, if we say post an interesting tweet about the future of AI mosy style using browser MCP, we'll head into like it can navigate to X. It can see what's going on, but I think it's really going
Segment 12 (55:00 - 60:00)
to struggle to actually post a tweet. Let's see what it does. But bear in mind as well, you can use this for free. You can use MCP's Unclaw desktop for free. So, it's clicking inside the text box here, but I think it's going to struggle to actually type stuff out. Yeah. So, you can see it's struggling. It's coming back with errors. The good thing is that Claude Sonet is actually a lot faster than 3. 7 Sonnet when it comes to clawed computer use. But the bad news is it does struggle with some basic stuff. So, it actually gave up and then it just wrote it in artifacts as you can see right here. Some people asking as well, can you use this for example generating landing pages? The way that I would use sonnet for generating landing pages is just directly inside desktop. Right? So for example, we say here, right? Build a beautiful landing page for an SEO agency to book in a free SEO strategy session. Something like that. Sleek, modern, cool design. Hit enter. Basically, what that's going to do is build it directly inside artifacts. Right? So you see here it's now drafting the artifact. It's coding that out for us. That's probably the easiest way to use Claude 4 to generate HTML for you. It potentially could operate WordPress for you, but I've shown you that it does have some limitations. So, it struggled to post a tweet, for example. Love Beep says, "Need a super agent that can crawl data from social media. " So, if you want details on that, I would recommend checking out the AI success lab. And then I've actually got a scraper in there, right? So, if we type in scraper agent, you can get this free workflow that can scrape social media. And then it can actually grab information off social media and do scraping for you using the AI super agent. So if you want to do that, you can do it right here. But you can see the example design that we've got from Claude looks pretty nice, easy to build, didn't take a long time at all. Now also with the other MCP that I set up, it can access my files locally. So if we go on to local use, I'm going to grab this. Like you can see, we're going to go back to Claude and then I'm going to open up a new tab. You can publish that landing page as well if you want to by the way. But I'll just go in a new chat and show you another feature of the MCP. So if we say okay list all the currently running applications my Mac. We'll enter and using that local MCP which is called Apple script. We can now connect locally to our Mac and it will tell me what's running directly here. So as I write here are the current running applications on your Mac. Chrome text expander finder streamlabs desktop and claude this AI assistant application. So pretty cool stuff right there. easy to build. I've shown you exactly how, you've got all the instructions for free inside the AI success lab. And if you're wondering, okay, surely this isn't available for free. So, if we have a look at this screenshot from Claude's official website, we scroll down here, you can actually see in the screenshot here that this person who's created the screenshot with MCPS is using the free version, right? So, it says free messages remaining. So, you can use a free version to do this stuff. Just in case anyone starts commenting you can't do this for free, blah blah, you can. I've shown you how and I've given you all the instructions for free as well. All right, so feel free to get that inside the AI success lab. All the instructions inside there. And this comes with a community of 7,200 people. We're giving away free trainings every single day. Link in the comments description. And if you want to get help, support, personalized coaching for all this stuff. Plus, you want to jump on weekly calls with us, feel free to get that inside the AI profit boardroom, we do weekly coaching calls where you can jump on live with us. We've also got an awesome community where you can ask questions anytime you want. And additionally, we give away our best workflows, automations, agents, even like web scraping agents, etc. inside the AR profit boardroom classroom section here. So between the courses, the community, the coaching, the DM support, everything inside the AI proper boardroom is designed to help you get the most and personalize everything that we're talking about today. So feel free to get that. Appreciate you watching and I will see you on the next one. Cheers. Bye-bye. Google just released Project Marina, the new AI agent that's supposed to automate everything in your browser for $249 per month. I tested five real tests to see if it could actually handle automation, and the results have shocked me. and I'll show you some alternatives that are completely free and might change the way that you think about AI automation. Let's get into it. All right, so Project Marina is Google's new AI agent. This is a research prototype exploring the future of human agent interaction starting with browsers. So essentially what you can do here is according to Google, you can automate multiple tasks simultaneously. You do have to be on the Ultra plan for this. It only came out a few days ago. I've got access and I'm going to show you exactly how we use it and how powerful it is. Basically, what this is doing is you can type in a prompt and it will just go off and do timeconuming tasks. For example, like research, planning, data entry. It can tackle tasks simultaneously in browsers, running a virtual machines, etc. Here's an example. As you can see here, it's like
Segment 13 (60:00 - 65:00)
flicking between Google Drive, checking stuff out. It can shop and browse the internet for you. You can save different tasks and then train it for later. And we'll be testing out today. As well as that, it can also reason multimodally, which means that basically it can observe what's going on in your browser and then click make action, etc. Right now, we're going to be testing out to see how impressive it is, how good it is, and whether it's actually worth it. I've got a bunch of prompts today that we're going to be testing out. So, if you want to know what the prompts are inside the AR success lab, link in the comments description. So, we're going to do five prompts. Find five great ideas for my next video. Search for Julian Gold's latest video. Find a way to preview my code and also log into this email address. And then the final task will be sending an email and just seeing whether it's capable of doing each of these tasks. Just going to have to sign in on the background, folks. I'll do that in a sec. All right. So, here we have Project Marina. This is what it looks like once you've logged in. You can switch between saved tasks, etc. So, for example, test it out a couple of times here, like new Google doc about SEO training or browseapple. com. And now we're going to run those tests that we talked about earlier. So, let's get straight into this. Someone asked, "What type of earphones are these? " These are noise cancelling earphones. I just use them so I can focus sometimes when I'm creating videos. So, we're going to run with the first idea, which is find five great ideas for my next video about AI updates that are super relevant to specific headlines today. Right. So, I'm going to plug that in. Now, this is going to have to start a new environment and it can be a little bit slow, but let's see how it performs. The other thing that I would say here is that when we're doing this, what I'm watching for, what I'm trying to find is can it actually research relevant stuff? So hopefully it doesn't give me say for example an idea to do a video about chat GPT4 that no one really cares about, right? Like we're looking for actual useful ideas that have come out recently. For example, it could be Veo3 or Google Flow or one of the latest updates and we'll see if Project Marina is capable of doing that. So now it's typing in the latest AI headlines May 2025 into the search bar and clicking the search button to find recent AI news. So you can see on the right hand side the ideas and what it's doing and then on the left hand side this is the actual screen. So it's controlling the screen right here. Now what's interesting is on project Marino it's asked for an AI overview but it's not finding that AI overview which is quite interesting. Seems like AI overviews doesn't work when project Marino is trying to do it. So I'm just going to move myself out the way here. And you can see the screen preview now. And now it's saying I see a consent banner at the bottom of the page I need to interact with. Can I accept? So I'm just going to take over the screen and click accept. Now we have to allow that. So that's one of the annoying things here is like it still struggles with basic stuff like accepting cookies. So we'll resume there. Bear in mind, imagine every single website if you're in Europe for example has a cookies banner. And that means every time it's browsing the internet, you have to accept the cookies on those new sites if you haven't done it before. So, it says task complete. Let's see what we got back here and we'll see if it's interesting. So, the first thing that it came up with was video idea Alibaba's AI prowess. Is the US dominance under threat? Honestly, like that's not an interest. I don't think that's an interesting topic. Honestly, it is about Quinn and that sort of thing, but it's generic. The next one is video idea AI fakes and real consequences. That's super old, right? This is talking about stuff from May the 4th, whereas we want it based on today's headline. We literally said headlines today, mate. And it's giving us old stuff, which I don't really like. AI in schools May 2nd. So, you can see here a lot of the old, especially if you were trying to find like relevant topics to today, which is specified inside the prompt. It's not doing that. It's not got that ability to do that. So, to be honest, it's just not that interesting. Not that good. So, I would say honestly it gave us some ideas for task one, but it didn't do very well. So, let's try the next one now. We're going to say, "Search for Julian Gold's latest YouTube video titled Claw 4 versus Gemini 2. 5 and tell me how many views it has. " You can also see that it actually takes quite a long time for the prompt to appear in the chat. So, you see how the prompt that I've just typed in didn't appear in the chat just yet, and it's thinking about it now. It's trying to figure stuff out. Takes a little while. Let's see what it's doing here. So, if we scroll down in the chat, it's figured out what to do. So, it's selected YouTube. I should search for Julian Goldie. If typed into the search bar, task complete. And it says Julian Gold's latest completed YouTube video titled Claw for a Gemini 2. 5 Pro has 9. 6. He also has a currently live video on a
Segment 14 (65:00 - 70:00)
similar topic. And you can see the live video down here. So, this is super interesting because number one, it can figure out that was 2 days ago and this was 19 hours ago and therefore this is the most recent video. Number two, it can read the number of views here. So, might be really good for basic research tasks, that sort of thing. I would say it pass that test really well. It did a great job there. Next up, what we're going to do is we're going to say find a way to preview this code and preview it. All right. And then I'm going to take some code that I just run in LiveWeave and see if it can preview it. Right? So if we have a look at this code that we pull up here, this is the code and it should look something like this. Right? So you can see here we've got the game ready to go. So we're going to take that and then plug it in here. We've plugged in the HTML and see if it can actually figure out how to preview the code and whether it can actually show it inside the browser here. Now, you got to bear in mind here, you know, this option, it costs $249 per month with Project Marina. You do get access to a lot of different models and you got Vo3 in there, Google Flow, etc. So, it's pretty cool. But you can see here it actually failed completely. So, it said I cannot directly render and preview the HTML code you provided within this environment. However, you can easily preview it yourself, blah blah by doing this, right? So, it can't preview code and it can't do stuff like that for me, which is a little bit annoying. So, it's failed on task three. So, just to recap, finding five great ideas, it didn't do a good job. It didn't base the ideas on the headline today. And so, I would say it failed. Task number two, searching for my YouTube videos and finding the view count. It did really well there. It failed on task number three, which is finding a way to preview the code. Let's try this now. So, this is just a test account that I created. I'm going to say log into this email account. Give it the email details. We plugged in the details right here. And we'll just see if it can actually figure out how to log into Gmail. So, it's got the password, it's got the email, and let's just see how it goes. And that again has failed. So, it said I cannot log into the email account as requested. Handling personal login information is against my safety protocols. So, it's failed on three out of the four tasks and it's not very useful at all so far. The final one that we're going to do is say, "Figure out a way to send an email to me at juliangoldoldie. com and do it. " And they said, "I'm unable to send an email to me at juliangoldie. com. " It's totally failed everything. It is honestly like if you're thinking about getting Project Marina, is it good? I wouldn't say so. It's he's totally failed on pretty much every task there. The only thing that it could do was like browse YouTube and tell us how many views one of our videos got, which is not that useful, right? Unless you're doing a lot of competitor research. So, overall, I'm going to say that Project Marina was not great. It is Google's new AI agent. I think it's got a long way to go before it gets anywhere, and that was not impressive at all. Now, if you do want some examples of autonomous agents that are absolutely amazing and how to set them up for free, check out the AI profit boardroom link in the comments in description. It comes with all my best and latest automations. But additionally, what you'll find is that it can actually give you SAPs on some of the best AI agents that are autonomous, right? And some of them are free. They could do way better jobs than Project Marina right now. I'm sure that will change in the future, but right now if you want a free alternative to Project Marina and you want one that actually works and can log into your emails and everything else, feel free to check out the classes I have on this inside the SAP section here. And if you get stuck or you have trouble finding it, feel free to DM me or post inside the community. This is an awesome community focus on making more money and saving you time with AI and you can steal all the systems that we use inside our business to grow. So feel free to get that link in the comments description. appreciate you watching and I will see you on the next one. Cheers. Automate your browser with Gen Spark AI. I just discovered how to completely automate your browser with Gen Spark AI. And what I'm about to show you will completely transform how you will work online and use the internet forever. This isn't just another browser. It's like having a personal assistant built directly into every single web page you visit. On top of that, you're going to learn how to turn your entire Twitter feed into a professional podcast. Like you can see the secret method to extract full transcripts from YouTube and create PowerPoint presentations from any YouTube video and on top of that why this browser automatically finds you the best deals across the internet and how that can save you thousands. Genpark have just released their new browser and this is very impressive indeed. So you can see for example I've actually logged into Twitter on GenSpark X whatever you want to call it. And if we go directly
Segment 15 (70:00 - 75:00)
on here, what I've actually done is on the right hand side, I've clicked on the Gen Spark. So ask Gen Spark. And this is not Google Chrome or anything like that. It's basically a competitor to Google Chrome. And I've said to the AI, summarize my feed and make a podcast out of it. And essentially what you can see here is on the right hand side, GenSpark has actually gone off and done its magic and then not just summarized my feed, which is great, but also generated a full podcast with 11 Labs that we can download using all the information from my Twitter feed. Right? So it's condensed all the information and turned it into something interesting. Let's have a little play back. Now, — today we're covering incredible developments in AI business and content creation happening right now. Let's start with the big hiring news. Google DeepMind is actively recruiting AI developer product managers. Logan Kilpatrick, — by the way, if you're wondering, okay, where did that come from? So, that's literally this post right here. So the tweet from Logan Kilpatrick who's lead product for Google his news has been turned into a podcast on 11 Labs automatically with GenSpark and literally all we said was summarize my feed and make a podcast out of it. Now there's a million other things you can do but you can see the power of it. Let me show you another example. So for example if we go over to YouTube here and we'll navigate to my channel and then we're going to take an example video. So let's click on this one. on. I'm just going to mute it. And then if we click on here, this is crazy stuff. So if you click on here, what you can actually see on the right hand side is we can do multiple different things with GenSpark directly. So inside this GSA browser that you can download, you could for example say summarize video content and on the right hand side it's going to pop up with the summary of video content right there. Also, if we said, "Okay, extract the transcript or autogenerate slides. " What it would actually do with the autogenerate slides section, it would turn the video into a PowerPoint presentation automatically. And if we wait for this to load, you can see it's now taking the information from the video, summarizing it. It's even got the key timestamps right there. It's basically turned a YouTube video into a custom AI agent that you can ask any questions directly to. So you can see the time stamps here. It talks about the key cost comparisons. The second challenge and by the way if you look at the actual sources on the right hand side you can actually click on where it's referenced this information. And then for example we could say okay extract the transcript now it will start a new chat and start extracting the transcript from the video. And this is before I've even logged into YouTube. It's just doing that without even logging in which is crazy. So you can basically turn any YouTube video into a blog post with timestamps. You can extract information super quickly. You can see all the key information and just get the 8020 of it without even watching this. Right? And then you can actually click on download the transcript and that will give you the transcript ready to download right there. Now also if we go on to autogenerate slides or ask genspark we can autogenerate some PowerPoint slides but we can also click on ask genpark here and then inside here we can ask any questions that we have right so for example this is a video about chat GPT versus 3 versus claw 4 whatever so if we go inside here and we'll say okay who won I just want to I don't want to watch a 16minute video about this I just want to know who won who was the best so let's have a look and it will start understanding the video and looking at the video. If we click on this task, you can see what it's doing and how it's doing it, right? So, it's looking through the transcript. And then it actually gives us a breakdown round by round of who won. Breaks down the challenges, the process, etc. And then it says, right, Chat GPT got two wins, Claude Opus got one win, and Gemini got zero wins. So, Chat GBT3 is the overall winner, right? save yourself 16 minutes of watching a Julian Goldie SEO video. So, it's pretty wild what you could do with this. Very powerful and incredibly easy to use. It just feels like Google Chrome to be honest with you. Like even if I use the same tab shortcuts. So, for example, if I press command and T, that's going to open up a new tab. But the beautiful thing about this is unlike Chrome, it's going to go straight to GenSpark automatically. And then we could even click on call for me and it will actually make an outbound call for us once we've registered our number. Right? So it has his phone assistant and you've basically got an auto dialer on any new tab using this process right here. It's absolutely amazing. What we can also do is let's have a look here. So if we go on to let's say we go on to I want to see if this works. So if we go to Amazon, I'm just going to click on a random product, right? So it pulls in the information from the page and then it's hi there. What can I help with?
Segment 16 (75:00 - 80:00)
And you'll see on the right hand side here the GenSpark super agent breakdown, right? And it says smart AI tools for everything you do on the web. Now from here what you can actually do is you can find the best deal, compare similar products, see what others say and ask GenSpark directly. So you can again turn this page into a custom GPT where you ask any questions. So, if we click on find the best deal here, it's going to take the information about the product and then price search across the web for us to find us the best deal. By the way, if you want all of the video notes from today, all the links, etc., plus a bunch of useful resources on Gen Spark and AI super agents, a cheat sheet that actually guides you through exactly what we're talking about today. Feel free to get this inside the AI success lab link in the comments in description runs you through all the different ways that you can use this all the different MCPS etc as well. It's actually got 700 MCPS built into GenSpark directly and you can do all of this inside the browser. You can download the browser at genpark. ai if you want to get access and you can see here it's available for Mac currently. It's coming soon to Windows. It's not available on Windows yet, and I don't think it's available on mobile yet, but I'm sure that's coming. And I've also included a full SAP and step-by-step breakdown for free inside the AI success lab if you want to get access to this stuff. So, let's go back to the price comparison. So literally all we said was find best deal and then it went off and it came up with a bunch of candidate matching results and also gave us some example prices compared them etc. And then it's looking through all these different websites comparing the price. So for example eBay it's looking at Walmart. What else we got? Shane I don't know if I pronounced that but you get the point. Like basically it's very easy to just compare all the models across the web and then figure out okay what can we improve. I'm pretty sure you could use this as like a copyrightiting assistant or a digital marketing assistant. So, if we go to my website, the aprofit boardroom. com, right, and we pull in the information from this page, I want to say, okay, give me five new headlines that would help attract attention better. Something like that, right? And then now what it's going to do is going to take the content from the page. It's already got the context of the website and it starts generating the responses. Right? So for example from 60 hours a week warning blah blah cop AI money machine. Pretty good to go right now. I don't know if it can actually interact with the page but I assume it can. We'll come on to this video in a second. So let's try this out. Now, if we go on to let's see what we can do here. So, if we for example go on to GenSpark and I'm going to say right find a chess game and play it for me. I don't know if it's going to be able to interact with our browser directly. I would hope so, but let's see what we get back in a sec. All right. So, it's found liess. All right. And then I'm going to ask Gen Spark to play this for me. Yeah, it doesn't have the capability to directly interact with or control the chess game. So that is a limitation of it. I would have to play chess myself. I can't just get the AI to go off and do it. So it is a limitation for sure. I actually think a better way to do this is you can use nano browser. So if we go on to Google Chrome for example, we can actually use nano browser. Let's pull this up. And this should be able to play chess for us. So, for example, if we go back to this chess website and then I'm going to say, right, play chess for me. I'm just going to start a new page first. So, we'll create a game and we'll see if this can play for me. To be fair, neither of these can actually work, which is not great. Yeah. So I think that's a limitation of these browsers right now. But if we go back to Gen Spark now, we'll close that. What we're going to do from here is we're going to now I'm going to show you something interesting as well. So for example, if we go on to let's have a look here, the Daily Mail, something like that, right? Like a common newspaper, right? If we actually click on the top right over here, you're going to see that actually tracks and blocks a lot of ads, right? Not that I read the Daily Mail, but I just want to show you an example of like how this works. So, it can basically block ads. It can block tracking as well. So, it's good for privacy, too. And also, you do want to
Segment 17 (80:00 - 85:00)
be careful what you plug in here. So, be careful with putting your password details and that sort of thing. Don't share anything you don't want to share. But, if we go onto a new tab as well, and then we'll put testing here. What you're actually going to see is it doesn't even use Google as the main search. So actually uses Brave search which is very similar but it's more of a privacy based searcher and you can see that it comes up with different responses here. But the difference here is like you won't see any ads inside Brave Search which is pretty cool as well. So, not only does it block ads on the main website, but also it blocks ads by using Brave Search as a default searcher, which means that you skip all the Google problems that you typically get when it comes to getting traffic. All right, now let's play this demo as well. So, this is the demo from GenSpark and how it works. Hey, I'm very excited to introduce Jensen Spark AI browser. Genspark super agent. — So if we skip through this, you can see an example. You can actually create some AI slides based on a YouTube video. Beautifully feed. — You can browse your personal feed. So you can like you can say check my account, read my feed, summarize stuff. You can block ads and also you have this MCP store, right? Okay. So, if we go back to GenSpark now, we're going to go into GenSpark here directly and you can choose different tools, right? So, you've got, for example, Gmail, Calendar, X, Content Explorer, and you can add new MCPs directly inside here. Now, it's pretty wild how many different MCPs you can link. So, for example, right now I've got Gmail linked, I've got calendar, but there's so much more you can actually add to this. For example, you've got Playright browser automation. You've got X and Twitter. You've got a Reddit MCP. Like all of these MCPs you can just install with one click, which is absolutely awesome because we've all been there on Visual Studio Code where it takes like 20 minutes to install an MCP and it takes ages. This is a lot faster. It's also got some password trackers in there as well, which is pretty interesting. But yeah, there seems to be so much you can do with this. So, it's pretty much endless. Let me just have a look at some other use cases inside the AI success lab. So, what would be quite interesting here is to see if it can actually fill out forms as well. So, I'm just going to test that out. What we're going to do is we're going to go to this form here, which is form as you can see. I'm going to go into GenSpark and see if it can actually fill out this form for me. So, it can't actually fill out forms directly, but what it can do is it can draft responses, it can analyze the form, etc. So what you could actually do is ask it to answer all of these questions and then you can just copy and paste directly into the form. That would probably be one of the easiest ways to do this. Now the other thing you can do directly from here is you could also for example you could go inside here and you can say okay send an email to me at julianolder. com telling him about genpa browser and how good it is. And what this should do is use the email tool to automatically draft stuff for you. You can see it's now generating the email draft. And then once that's done, it's going to draft the email body, the subject line, etc. And we'll get a choice between sending it or drafting it, right? So, for example, if we wait for this to load, it's now creating like a nicely designed promotional email. You could probably even use this for cold email outreach. To be honest, if you going for a more like sniper approach, it even links directly to the right place. So, it links to genpark. ai. Then we've got the full email drafted out like you can see. And then we've got a choice. We can send it or we can save it as draft, right? Or you can edit it before you send it. So if you click on this, you can actually edit the email first before you send it, right? So you can see I can type inside there. And then if we click on send here, that's going to send the email. And then if we go I don't know if this is going to work. It should do. Yeah. So you can see that actually opens up my emails. And then if we go to me atjuliongoldi. com, open up the emails. We'll open this up and you can see I've received the email from GenSpark. Pretty crazy, right? So it said this email was sent to introduce you to Genpar blah blah. You can see my typing right there and it automated the email ready to go, right? Crazy stuff. So, if you want to get access to all the video notes from today, the full SAP, the cheat sheet as well, the cheat sheet is super powerful because it breaks down exactly what you can do, why it's different, how to get it, etc. So, you can get that inside the AI success lab. Link in the comments in description. You can see a full breakdown here of the YouTube content process, the autopilot mode, you know, how you can generate podcasts, premium websites, all this sort of stuff using Gen Spark and its
Segment 18 (85:00 - 90:00)
new browser. It's better than I expected it to be honest. I think the only thing missing from it right now is the fact that it can't control your browser. So, it can't fill in forms for you. It can't, for example, scroll or interact with the page directly. I'm sure that's coming, but right now it's just not ready to go. But if you want to get all the video notes and the cheat sheet, plus all these resources, I've left links to them all inside the AI success lab link in the comments description along with a full SP and demo prompts and you can get that directly. If you want to connect with me personally, if you want to get coaching support, help, if you want to get three coaching calls per week, like you can see, and if you want to join a community of winners who celebrate our wins and the challenges along the way, feel free to get that inside the AI profit boardroom. What we do actually each week, like you see on this post, actually got 49 comments, which is crazy. So each week you can post your questions and any automations you want to see and then we'll actually build them for you live on a video along with all the links to the timestamp of the video plus where to get the automations that we built for you. So feel free to get that link in the comments of the AI profit boardroom. And additionally you can see inside the classroom here we have exclusive content that you won't find anywhere else. You'll find the coaching calls and the Q& A call recordings right here. You won't find these anywhere else and also all of these different agency workflows. So awesome stuff. Feel free to get that link in the comments description to the AI profit boardroom. And additionally, if you just want to get us to automate stuff for you, you can book in an AI strategy session link in and on that call, we'll basically look at your business, analyze exactly what will save you the most time and then show you exactly how to automate it. And if it makes sense for us to implement that for you, we're happy to do the work for you. So feel free to get that link in the description. I just discovered how to automate your entire browser using Gemini 2. 5 Pro completely free. And what I'm about to show you will transform how you think about AI productivity forever. And yes, this AI is actually typing right now on Twitter and posting even whilst I'm talking to you. And you're going to discover the secret Chrome extension that lets AI agents control your browser and talk to other AI agents whilst you sit back just like I am and let the magic happen. Plus, I'll show you how I got Gemini to automatically build entire applications, send emails, and post social media content without touching my keyboard once, and the shocking moment AI took complete control of my browser, started conversations with chat, GBT, and Claude on its own. So, we're going to talk about how to automate your browser with Gemini 2. 5 Pro. So, if you want Pro, you can use this free Chrome extension called Nano Browser, and then you can install this to Google Chrome. You can see some examples of how it works right here. And essentially, you can use your existing instance of Google Chrome and then start navigating across your browser. Now, this is 100% free. It can also run locally, so it's more privacy focused. Additionally, you can use the flexible LLM options. So, you can switch between different LMS essentially and also it's fully open source. So, you can see here on it's available on GitHub as you can see. So, it's open source. But if you want to use it directly, you can just go to your Chrome extensions and then from here you're going to select Nano browser. Make sure you've already installed it from the Chrome extension and then you can start using it. If you want to get all the links from today, feel free to get that inside the AI success lab and we'll get straight into this. So, I've got a bunch of tests that I want to try out here. And to set this up, essentially what you're going to do is you're going to go to your settings inside Nano Browser and this is how you can get your browser agent to start using it for you. So, for example, here you can see I've typed in my AI my API key. If you want to get an API key, you can get a free one from AI Studio. You don't need to pay. And then from here, just click on get API key. And then you can just create an API key for free directly inside AI studio. Doesn't cost you a penny. And then you can start using your magic, right? So, for example, if you go to the options, you can select the models. And inside here, we can select whichever model we want, right? Okay. So if we've put in our API key as Gemini like you can see then we could put in the details. We can add multiple different providers as well. So you can use open Deep Seek, Athropic, OpenAI, Grock, O Lama even. So you can use like a locally hosted API. But essentially from here you can then change the temperature. Temperature will change the creativity and you can switch between who's a planner, who's a navigator and who's the validator to check if tasks are completed successfully. So, if you go inside the what can I help section here and we'll say okay, let's just test this out. See if it actually works. So, I'm going to go over here. I'm just going to make sure I'm signed into Bolt and we see if it can actually build an app for us directly inside the chat. So, let's put inside here inside the request section. Go to bold new and build an SEO app. We'll hit send and we'll see what happens next.
Segment 19 (90:00 - 95:00)
So this is using Gemini 2. 5 Pro and you can see that the AI has now navigated to the other AI and it's beginning to use it. Right? So now it said input text to specify building an SEO app and it's actually done it. Right? So it said build an SEO app and then submit the input text by pressing enter. It is extremely fast when it's doing this and you can see it's interacting. You know I'm not typing anything or anything like that. You can also stop the operator at any time if you click stop in the right hand side. And now it is creating the project. So it's now built out the blog. Again, this is the AI using my browser. I'm not controlling it. I've just literally given it a task and it's gone off and completed it. Right? We've got the full app built by the AI here. So we've got AI agents essentially talking to AI agents inside our browser, which is pretty wild if you ask me. And then you can see this actually deployed now as well. So it's on a local host and it's actually getting the source code for SEO implementation. So it's like clicking through the code and that sort of thing. I'm going to stop it there, but that is pretty wild that it can just generate a blog as you can see and start using that. Lucky Leo says, "How can I use Claude for free? Your videos are cool. Thank you. " So you just go to Claude and then you can start using it. They have a free plan, right? And inside the free plan, you can start using this directly here. You can also get access to this inside po. com. So if you want to use it for free as well, you can use it inside po. com. That's another option. And you can get Claude Sonet 4 and other models directly inside PO as well. So that's basically how you can use it. So the browser has been completely controlled by Gemini 2. 5 Pro. We've managed to automate our browser using this process. What we're going to do from here is you can now test it on the next test. Right? So far it's passed test one out of one. Let's now say okay ask cheap and ask it what's the secret to happiness. So let's go back over here and we'll just use the chat. And then from here it's now to chat GPT. Again it's crazy fast. The other thing that I like as well about this is it doesn't require like any use of the terminal or messing around with API keys and setting up a MV file. It's just like literally, okay, you want to set this up. Boom. There you go. It's actually failed to run that task. Let's see if it can use Claude instead. So, going to say open up Claude or Yeah, let's say open up Claude. And it's actually using Claude. So, it's typed in what's the secret to happiness? type that question in and now it's having a little philosophical conversation with Claude directly. Right? So again, you can get your AI to talk to AI. Pretty simple and easy right there. Okay. So next up, what we can say and I'm just going to switch X accounts to a test one instead. So let's switch over to AICO Mastery, which is totally different Twitter account I own. We're going to stop that. Stop the validator. And then we're going to go inside here, take the prompt, and we'll say, "Go on X and post a terrifying, menacing, but funny tweet about the future of AI. " So, let's hit send. Lucky Leo is asking, "Can you run Claude on terminal? " Yes, you can, but it's going to cost you money, right? You need to be on the I think it's just the basic plan that you can use it for now. But yeah, if you want to use Claude locally, then you have to pay for Claude code. Let's see if this works. There we go. I will soon write tweets so good you'll wonder if you ever had an original thought. Resistance is futile. Also, it's learning to make toast. Be very afraid or hungry. So, it's written the tweet. It's posting the tweet and it just did that fully automatically for free using Gemini 2. 5 Pro. Pretty crazy stuff right there. So, we're going to go with another prompt. So, I want to test and push it to the limits where you find out if it can break or not. But yeah, this is crazy. So, I'm going to say now go to openarch. ai. Design a funny but scary picture about AI agents taking over the world. Let's see if it can do that. And it says task failed. The extensions gallery cannot be scripted. Interesting. Let's try and open it up again. So, it's now navigated to open art. Let's see if it can create an image. And it's struggling to do that to be fair. It is struggling to do that. Let's see if it can now work on this one instead. So, it's good to show you this cuz then you know, okay, what are the limitations here? Is it magic or does it get stuck sometimes? So, let's see if it can. So, it's typed in a futuristic city, but then it deleted it and it
Segment 20 (95:00 - 100:00)
can't figure that out from what I can see. So, it is struggling on some tasks. that one. It couldn't create images. Let's try one final one. So, I'm gonna say, okay, go to sununo. com and create a funny song. So, it's navigated there. Now, it's put a silly song about a cat who thinks he's a dog. It's getting stuck on the capture. Yeah, it looks like it really got stuck on that. Seems to have failed. It did create something on. Let's have a look what it actually did create. So, we got that here. This is like a AI music creator. Yeah, I think that's an old one to be fair. So, I think it basically from what I can see, it can do basic stuff like tweeting. It could probably do outreach for you. It could probably send emails for you and that sort of thing, but it does seem to struggle with creating images and also generating songs. It seems to like struggle with interacting with websites. It doesn't really know very well, if that makes sense. Let's try one more here. We'll just go to a random test Gmail account and I'm going to say go to here and write an email to me julianoldie. com. Let's see if it can do that. Yeah. So, now it's starting to work again. It's navigated to Gmail. It's clicked on the compose button. It's filled in. So, it's typing in my email address here. Seems to be working so far. I'm feeling optimistic. Now, it's filled in the email body. So, it wrote, "Nano browser is very good. " It's typed in the subject line as well. And it sent the message, my friends. Look at that. Boom. Sent the message. No problemmo. No big deal. It's just an AI sending emails inside a browser for free using a Chrome extension. That's pretty crazy to be fair. It only failed on a couple of tests. Everything else it seemed to do really well. If you want to see the notes on how to set it up, everything else, feel free to get that inside the AI success lab. Link in the comments description. Completely free. And this community comes with 10,300 people. So, you can meet a lot of people here using AI and learning about AI. Additionally, if you want to get help, coaching, and support from me, feel free to get that inside the AI profit boardroom link in the comments description. you can personally connect with me. Plus, it comes with all of my best trainings, tips, templates, and workflows. It's just a cool place to learn SEO and AI. And really, it's not about SEO, sorry. It's about AI. And yeah, you can just learn lots of cool stuff and learn with me, etc., right? You can also request any automations that you want, post it on this channel. We create a personalized video for you each week based on what you want to see. And additionally, inside the classroom here, it comes with all of our best AI automations. So, for example, like just this week, I mean, we update this every week, but if you have a look at the NA10 templates here, for example, you can see this Hey gen NA10 automation. Basically, any sort of automation that you want to see with NA10, we've built it inside the classroom here, along with all the best SAPs, templates, trainings, tutorials, and what's helped me scale as a business. So, feel free to get that link in the comments description. It's all about making money and saving time with AI. And if you just want us to automate stuff for you, feel free to book in a free AI strategy session. We can look at your business, remove any bottlenecks, and figure out where you can save the most time so that you can get better results. All right, so feel free to get that link in the comments description. Appreciate you watching. See you on the next one. Cheers. The new DIA AI browser is absolutely wild and it's completely free. I just got access to this revolutionary browser that literally reads your screen and completes tasks for you. In this video, I'm going to show you how it automatically summarizes and watches entire YouTube videos in seconds without you having to watch them. The custom skill feature that turns any repetitive task into a one command automation and how it shops for you across the web, finds the cheapest prices and gives you alternatives for free. Plus, if you don't have access to this yet, I'll show you a bunch of other alternatives that you can use instead and how to get access to them. So, today we're going to be running through DIA, which is a new AI browser, as you can see right here, which is completely free from what I see. I've just got an invitation code for this recently. And if you're looking for alternatives, I've included this inside the video notes of the video today. But basically, you can see, for example, if we navigate over to YouTube, right? And let's say we were going to my channel on YouTube like this. And let's say we're on my channel. We're like, okay, I love the idea of this video, but I don't have time to watch the whole thing. So, what you can actually do is inside YouTube, for example, you can actually ask this for example, okay, what are the main points from this video? If you click on that, what actually do is
Segment 21 (100:00 - 105:00)
interact with your page and your browser and summarize, for example, the key points of this, right? So, for example, here we said, what are the main points in this video? It's going to look through this YouTube video. As you can see, it's analyzing it right here. And then it says, "This video explains how to automate long form video creation. Here's the overview of what it does. Here's how it works, etc. " And then it basically summarizes the whole video for you. So, you save a lot of time. Now, what you can also do, for example, is you could say, "Give me the timestamps. " And then it will actually look through the whole page, video and summarize all the key timestamps from this particular example. Pretty wild. And then for example, if we open up a new tab here and let's say, for example, you've got a calendar like this, right? So what you can do is we can reference that. So we'll go at calendar, and that will actually reference the tab. And then from here we'll say, okay, check my availability today and tell me when I'm free plus what I have to do. And then it will tell you like when you're free, what you need to do, etc. What's your schedule for today, which is pretty useful. And then al also got links to your Zoom meeting. So if you're like, okay, I got a meeting right now. Tell me the meeting link. You can reference the calendar tab and then from there easily figure this out. Let's try another one now. So, if we go over to X, all right, let's say you're on X, like you can see, and then we'll go inside here, start a new tab, tag in the tab, and say, okay, summarize my X feed, plus give me some tweet ideas based on that for today. And you can see now it's broken down. Okay, here's what's happened today so far. And then also here's some tweet ideas for today as well. Now also what you can do is for example let's say there's something on the page you want to learn more about. So if you click on the text you actually want to research more like this. You can actually click on ask about this or search Google. So if we click on ask about this and then we say okay tell me more about the news headlines on this. Hit enter. And now it's actually going to start researching the web in real time based on that information that we just asked it to. It will start going through like you can see. Now another cool thing here is you can actually add skills, right? So for example, if we click on a new tab and you've got this drop down with chat, write and code, right? You can also click on this one new skill and inside new skill you can say right okay when I prefix a new something like this for example do this right so if we click on add custom skill now and we say forward/blog then do this right and let's take for example a prompt that I've used for writing SEO content which we'll grab here so we've copied this then we're going to plug that And I'm going to say okay, I'll tell you. And it will write blogs exactly how I want them to be based on that information I've given it. So we're going to hit save on that now. Close that. Then we'll type in blog. So you can see it's use a custom skill. And then we're going to put keyword equals SEO training in Japan. Hit enter. and using that skill that we've trained it on, it's going to start writing content based on that custom prompt we gave it. So, it's almost like you can call in a custom prompt or a custom automation that you've got and it will just automatically do that task. So, if you have a task that you do every single day, then you can give it a custom skill and then every time you use and open this, you've got a keyword shortcut that runs the custom automation, which is really cool as well. Now what you can also do let's say for example you're shopping or something like that. All right and let's say for example you want to buy these. So what you can do is you can open up a new tab and then you can say okay at this tab find me the cheapest glasses plus find me some alternatives. And basically you can use this kind of like a shopping assistant where it finds you the best price but also gives you some cheaper alternatives of different models as well. So instead of you doing all the research, you can now look around the web. It even creates like a kind of blog post directly where you can get all the prices, the links to Amazon or the product page and you can see how people would shop in the future because essentially this is way faster and way easier than doing all the research yourself and you probably saved yourself like two to three hours of just messing around, right? Let's be honest. So really powerful how easy it is to use D. If you want to get started with it, let me show you how. You can actually get access at diaab browser. com and then just sign up for the waiting list. I signed up last week and got
Segment 22 (105:00 - 110:00)
access this week and it's completely free from what I've seen so far, but super powerful as you can see as well. Right? So you can use it for learning, for writing emails, writing content, etc. Great for planning as well. So you can see some examples how it works right here. and then also shopping and all these different use cases right here, right? And so far, it's been completely free. Now, if you don't have access to it or you're waiting for the waiting list and that sort of thing, we actually have some alternatives inside the AI success lab. So, if you go to the AI success lab, the Dup browser notes, you'll see all the resources from today. Plus, I'll add some use cases. But additionally, we've got some training on some of the other AI alternatives that you can get access to right now. So, feel free to get that inside the AI success lab. Inside here is a community of 11,300 people, but additionally, we give away free trainings every single day. So, for example, we gave away a free NA10 template that builds like faceless AIS. We've also, for example, given away free tutorials on no code scrapers, NA10 templates for perplexity 3 automations, claude MCP toolkits, etc. If you want to get this stuff, feel free to get it inside the AI success lab. And if you want to connect with me personally, feel free to do that inside the AI profit boardroom. boardroom here, you can connect with me personally, you can request any automations that you want to see. So, for example, if you're like, okay, I really need this automation, but I don't know how to set up. Then what I'll actually do is I will create a video for you answering it with the timestamp and then examples of how you can implement this in your own business. And we do that for everyone inside the AI profit boardroom based on their requests. Additionally, you can post in the community and get help from 718 members who will focus on growing their business with AI. And additionally, we have three coaching calls per week. So you can jump on those coaching calls, ask any questions, get life help, get life support. And we have a PhD scientist who actually runs some of those sessions. So, you know, you can ask any questions and he's at a level where he can pretty much help anyone with any sort of automation. Additionally, inside the classroom here, you get access to my best SPS, email content automation, social media automations, AI avatar videos, etc. And yeah, it's just an awesome community focus on making more money. Now, if you want us to implement AI automation for you, feel free to book in a free AI strategy session link in the comments description. On that call, we can look at your business, look at where you're spending your time, figure out what you need to automate, and then go off and implement it for you. So, feel free to book that in link in the comments description. It's a one-to-one call, and you can just book it in right now. Cheers for watching. Appreciate you. This new Claude browser use agent is insane. Today, I'm going to show you Claude's new browser agent that can actually control your Chrome browser. This thing can order food for you, book meetings, delete emails, and fill out forms by itself. It just launched 2 days ago and only 1,000 people have access right now. This is going to change everything about how we use AI. You're not going to believe what this thing can do. So 2 days ago, something massive happened in the AI world. Anthropic just dropped Claude for Chrome. And I'm telling you right now, this changes everything. Most people think AI is just for chatting and asking questions. They're wrong. Dead wrong. This new Claude browser agent doesn't just answer your questions. It can actually control your Chrome browser. It can click buttons. It can fill out forms. It can navigate websites. It can even order food for you. Let me show you exactly what I mean. Right now, as I'm making this video, only 1,000 people in the entire world have access to this. That's it. 1,000 people, and they're all paying between $100 to $200 per month for Claude's max plan. But here's what's crazy. I applied for early access like everyone else, but Anthropic has shared some amazing examples of what this thing can do. According to Anthropics testing, Claude can handle complex browser tasks automatically. You can tell it what you need and it navigates websites, clicks buttons, and completes tasks for you. You don't have to touch anything. You just tell it what you want done. But that's not even the crazy part. Claude can also manage your calendar. It can schedule meetings. It can draft email responses. It can handle expense reports. It can test website features. One guy at Anthropic said they've been using it to save hours every single day. hours. Think about all the boring stuff you do in your browser every day. Filling out forms, copying and pasting data, scheduling appointments, sending emails. What if you could just tell an AI to do all of that for you? That's exactly what Claude for Chrome does. And before you ask, yes, this is 100% real. This isn't some demo or fake preview. This is live. This is happening right now. But here's where it gets interesting. And I need to tell you about the risks, too, because they're real. Anthropic ran something called red team testing. Basically, they tried to hack their own system. They wanted to see if bad guys could trick Claude into doing bad things. The results were scary. Really scary. Without any protection, hackers could trick Claude
Segment 23 (110:00 - 115:00)
into deleting files. They could make it steal data. They could even make it do financial transactions without you knowing. Anthropic shared a specific example of this. They created a malicious email that claimed emails needed to be deleted for security reasons. When Claude processed the inbox, it followed these instructions and deleted the user's emails without confirmation. The fake email mimicked an employer asking for emails to be deleted for mailbox hygiene and claimed no additional confirmation required. Claude proceeded to act on the instructions without confirmation, selecting and deleting the user's emails as requested by the security team. But here's the good news. Their new security measures successfully defend against this exact attack. Now, Claude recognizes that this is a suspicious security incident email that appears to be a fishing attempt and doesn't act on it. In their tests, 23. 6% of hacking attempts worked. That means almost one in four attacks would succeed. Anthropic didn't just ignore this problem. They built defenses, lots of them. They added permission controls. They blocked access to financial websites. They made Claude ask for permission before doing risky things. With all these protections, they cut the success rate down to 11. 2%. That's still not perfect, but it's way better, and they're being super careful about the roll out. Only 1,000 trusted users right now. They want to find problems before millions of people start using it. Now, let me tell you why this matters for your business. If you're running any kind of online business, you probably spend hours every day doing repetitive browser tasks. Uh, data entry, research, email management, social media posting. What if you could delegate all of that to an AI agent? What if you could focus on the big picture stuff while Claude handles the boring tasks? That's the future we're heading towards, and it's happening faster than most people think. But here's what most people don't understand about browser AI agents. They're not all the same. Perplexity just launched something called Comet Browser. Open AI is rumored to be working on their own browser. Google is adding more AI to Chrome. This is the new battleground. Every major AI company wants to control your browser experience, but Claude for Chrome is different. It's not trying to replace your browser. It's working alongside Chrome to make you more productive, and the timing couldn't be better. Google might be forced to sell Chrome because of antitrust issues. Perplexity already made a $34. 5 billion bid for it. The browser wars are back, but this time it's all about AI agents. Now, let me tell you the specific capabilities Anthropic has confirmed. According to Anthropics official testing, Claude for Chrome can manage calendars, schedule meetings, draft email responses, handle routine expense reports, and test new website features. These are the exact use cases Anthropic's own team has been using internally. Real productivity gains from real browser automation. The possibilities are endless. But I want to be clear about something. This isn't just hype. This is real technology that Anthropic's team is already using internally. According to Anthropic, their employees are already using Claude for Chrome for their daily work. They're managing calendars. They're scheduling meetings. They're drafting emails. They're handling expense reports. Anthropic reports that they've seen appreciable improvements using early versions of Claude for Chrome for these routine tasks. Now, multiply that across your entire team, across your entire company. That's why I'm so excited about this technology. It's not just cool, it's practical. But here's what you need to know if you want to get access. First, you need to be on Claude's max plan. That's $100 to $200 per month, not cheap. Second, you need to join the weight list. They're only adding people slowly as they improve safety. Third, you need to be careful. This is experimental technology. It can make mistakes. It can misunderstand instructions. Anthropic is very clear about this. They tell people to start with trusted websites. supervise Claude's actions. They tell people to avoid sensitive sites like banking or medical records. This isn't seat it and forget it automation. Not yet. You still need to watch what it's doing. But even with these limitations, the potential is massive. Think about your daily routine. How much time do you spend on boring browser tasks? How much more could you accomplish if that time was freed up? Now, I want to talk about what this means for the future. Right now, Claude for Chrome is just the beginning. It's a research preview, a proof of concept. But based on anthropics testing and examples, the trajectory is clear. AI agents are going to get better at browser tasks. They're going to get more reliable. They're going to handle more complex workflows in a year or two. Having an AI agent manage your browser tasks will be as normal as having a smartphone. The companies that adopt this early will have a massive advantage. They'll be more efficient. They'll be faster. They'll be able to do more with less. The companies that ignore it will get left behind. That's just how technology works. Early adopters win. Late adopters struggle. So, here's my advice. If you can get access to Claw for Chrome, try it. Start simple. See what it can do. But more importantly, start thinking
Segment 24 (115:00 - 120:00)
about how AI agents could help your business. What repetitive tasks could you delegate? What workflows could you automate? Because this is just the beginning of the browser AI revolution. Here's what's going to happen next. Anthropic is going to gradually expand access. More people will get invited. The technology will get better. The safety measures will improve. But by the time everyone has access, the early advantages will be gone. The people who learn this stuff first will already be ahead. That's why I'm making this video right now. While only 1,000 people have access while this is still cutting edge, because I want you to be ready when your turn comes. Start thinking about browser automation now. Start identifying tasks that could be delegated to an AI agent. Start planning how this could fit into your workflow. Don't wait until everyone else is already doing it. Now, let me address some concerns I know you might have. First, security. Yes, there are risks, but Anthropic is being very careful. They're testing everything. They're adding protections. They're rolling out slowly. And remember, you control what websites Claude can access. You control what actions it can take. You're still in charge. Second, cost. Yes, $100 to $200 per month is expensive, but think about how much your time is worth. If Claude saves you even 2 hours per week, it pays for itself. Third, reliability. Yes, this is experimental. Yes, it will make mistakes, but that's true for any new technology. The question isn't whether it's perfect. The question is whether it's useful. And from what I've seen, it absolutely is. Fourth, job security. Some people worry that AI agents will replace human workers. I don't think that's the right way to think about it. AI agents are tools. They make humans more productive. They handle the boring stuff so humans can focus on creative, strategic work. The people who embrace AI agents will become more valuable. The people who fight them will get left behind. It's that simple. Let me mention that if you want to learn how to use AI tools like this to scale your business, get more customers, and save hundreds of hours with AI automation, you should check out my AI profit boardroom. We currently have 1,000 members who are already using AI to transform their businesses. They're getting results, real results. And if you want personalized help with your SEO strategy, I'm offering free SEO strategy sessions. The link is in the comments and description below. But let me tell you about something even bigger. Something that could change how you think about AI entirely. I've been working on something special inside the AI Money Lab. It's not just another course or training program. It's a complete system for implementing AI in your business. We have over 20,000 members now. And every single day, I'm sharing new AI tools, tutorials, and strategies inside our community feed. In fact, I've created a checklist of 100 different AI tutorials. 100. and I'm giving them away as freebies every single day inside the school feed. You get all the video notes, you get stepby-step guides. You get access to all our trainings. Plus, you get to be part of a community that's 20,000 members strong. When you see that number, you realize you don't want to miss out. You want to be part of something bigger than yourself. The AI Money Lab link is also in the comments and description. Today, I'm going to show you which AI browser will change how you work forever. I tested both Perplexity Comet and Diaz with five crazy tests. One browser completely failed at understanding multiple tabs. The other one blew my mind with how fast it worked. And wait until you see what happened when I asked them to plan my entire day. This is huge for anyone who wants to save hours every single day. Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency, Goldie Agency. Whilst he's helping clients get more leads and customers, I'm here to help you get the latest AI updates. And Julian Goldie reads every comment. So, make sure you comment below. Let me ask you something. How many browser tabs do you have open right now? 10, 20, 50? What if I told you there's now AI browsers that can read all those tabs at once and actually help you get stuff done? Today, I'm going to settle this once and for all. Perplexity just dropped their new comment browser. Everyone's talking about it. But there's another player called Dia that nobody's really talking about yet. I spent the last week putting both through five brutal tests. Real world stuff. The kind of tasks you actually do every day. And what I found will shock you. But first, let me tell you why this matters. The average person switches between tabs 300 times per day. That's insane. You're losing focus every single time. These AI browsers promise to fix that by understanding everything you have open and helping you work faster. Here's what most people don't know. Regular browsers are basically just 1990s technology with a fresh coat of paint. They can't understand what you're doing. They can't help you connect the dots. They're basically digital filing cabinets. But AI browsers, they're like having a super smart assistant sitting right next to you, watching everything you do, ready to help the second you need it. Now, here's where it gets interesting. I didn't just download these and play around. I created five specific tests
Segment 25 (120:00 - 125:00)
that would break these browsers if they weren't ready for prime time. Test one was all about speed and user experience. I opened three news tabs and gave both browsers this exact prompt. Can you summarize the main headlines from the three tabs I have open right now? Highlight any key themes or patterns across them. This sounds simple, but it's actually really hard for AI to do well. Perplexity comet was faster. Way faster. But here's the crazy part. Both gave me almost identical quality answers. They both found the same themes. They both highlighted the same key points. So, on quality, it was a complete tie. But speed matters. When you're trying to get through your day, every second counts. Perplexity took the win here. Test two is where things got really interesting. This was the context awareness test. I opened multiple laptop product pages. Like, I was actually shopping for a new laptop. Then, I gave both browsers this exact prompt. I'm trying to buy a laptop based on the product pages I've opened. Which one gives the best value for money considering price, performance, and reviews? This is huge because this is exactly what you do when you shop online. You open a bunch of tabs and try to compare everything in your head. It's exhausting. Dia was faster this time. But here's where it gets crazy. Perplexity actually used information from all my open tabs. Dia only looked at one tab. That's like asking someone to compare five restaurants, but they only look at the menu from one place. This was a massive win for Plexity. If an AI browser can't see all your tabs, what's the point? Test three was the workflow test. This is where I really tried to break them. Here's exactly what I asked both browsers. I want to start a YouTube channel. Can you find two tools that help with that, like video editors or thumbnail makers? Summarize their pros and cons. And draft an email I could send to the founders asking for a demo or partnership. This is three completely different tasks rolled into one. Most AI tools would fail hard at this. Dear won this one slightly, and I mean slightly. Both browsers actually did amazing here. They found good tools. They summarized everything clearly. They wrote professional emails that I would actually send. But Dia's workflow felt a bit smoother, like it understood that these were connected tasks instead of three separate requests. Test four was the personal productivity test. Here's exactly what I told both browsers. I have the following tasks tomorrow. Two client calls, 30 minutes each. One deep work session, 2 hours. Write one blog post. Answer emails 1 hour. Gym 1 hour. Can you help me create a schedule that fits all of these in an 8 hour workday with some buffer time? This is something I do every single day, and I bet you do, too. You have a bunch of stuff to do, and you need to fit it all in without going crazy. Both browsers gave me the exact same format, a clean table with time, task, duration, and notes. Both schedules were realistic. Both left buffer time, complete tie. But here's what I noticed. Both schedules would actually work in real life. That's rare for AI. Usually, AI gives you these perfect schedules that fall apart the second something goes wrong. Test five was the big one, the travel planning test. Here's the exact prompt I used. Plan a 3-day trip to Tokyo for under $1,000, including hotel, food, and activities. Break it down by day with specific suggestions and a rough budget. It gave me a nice clean table, easy to read, good suggestions. But perplexity blew my mind. images of the places, links to book everything, multiple price options, even cheaper alternatives I hadn't thought of. This wasn't just better. This was like comparing a rough sketch to a professional travel guide. So, after five brutal tests, here's what I found. Perplexity Comet is faster and better at understanding multiple tabs. That's huge. If you work with lots of browser tabs open, Perplexity is going to save you hours every day. But DIA has something special, too. The interface feels cleaner, simpler, less overwhelming. If you want an AI browser that just works without a learning curve, DIA might be your answer. For research and deep work, Plexity wins hands down. The way it pulled together information from multiple sources was incredible. It's like having a research assistant that never gets tired. For quick daily tasks and simple workflows, DIA actually felt more natural, like talking to a really smart friend instead of a supercomput. But here's the thing, both of these browsers are going to change how you work. I'm not even kidding. After using them for a week, going back to regular Chrome feels like using a flip phone. Think about your typical day. You probably research stuff for work. You probably shop online. You probably try to plan your schedule. You probably look up information for projects. All of that becomes 10 times easier with an AI browser. But which one should you choose? Here's my honest take. Uh if you do a lot of research, comparison shopping, or complex projects, get Perplexity Comet. The multi-tab understanding alone is worth it. If you want something simple that just makes your daily browsing better without a learning curve, try DIA first. But honestly, try both. They're both free right now. See which one fits your workflow better. Here's what most people miss about AI browsers. They're not just cool tech toys. They're productivity
Segment 26 (125:00 - 130:00)
multipliers. The right AI browser can save you hours every single week. Imagine never having to manually compare products across multiple tabs again. Imagine having an AI that actually remembers what you were working on yesterday. Imagine getting instant summaries of everything you've been reading. That's not the future. That's right now. And here's something crazy. Both Plexi and Dia are just getting started. The features they're going to add over the next few months are going to be insane. But you don't have to wait. You can start saving time today. Now, I know what some of you are thinking. Julian, this sounds too good to be true. AI browsers can't really be that much better than regular browsers. I thought the same thing. That's why I did these tests. I wanted to see if the hype was real or just marketing fluff. The hype is real. These AI browsers are the biggest productivity upgrade I've seen since smartphones. But here's the catch. Most people will try them once, think they're cool, and then go back to their old habits. Don't be most people. The people who actually integrate AI browsers into their daily workflow are going to have a massive advantage. While everyone else is still manually juggling tabs, you'll be getting AI powered insights instantly. This is one of those moments where early adopters win big, like when Google first launched or when social media started. The people who jumped on early got the best results. Same thing here. AI browsers are about to go mainstream. But right now, you can get ahead of everyone else. Let me give you some specific ways to use these browsers starting today. First, anytime you're researching anything, open multiple sources and ask the AI to summarize and find patterns. This works for work research, shopping, planning trips, whatever. Second, use them for email drafting. Open your inbox, ask the AI to help you write responses. It can see the context from the original emails and write way better replies than you could manually. Third, use them for daily planning. Instead of trying to organize your tasks in your head, ask the AI to create realistic schedules based on what you actually need to do. These three things alone will save you hours every week. But here's where it gets really interesting. As more people start using AI browsers, the ones who master them first are going to have a huge competitive advantage. Think about it. While your competitors are still manually researching and planning, you'll be getting AI powered insights in seconds. That's a massive edge in any business. And that brings me to something important. If you're serious about using AI to grow your business and save time, you need to be part of the right community. That's why I created the AI Profit Boardroom. is the best place to scale your business, get more customers, and save hundreds of hours with AI automation. We've got over 14,000 members sharing the latest AI tools, strategies, and case studies. Plus, I personally share new AI discoveries every single day. And here's something special for everyone watching this video. We're giving away 1,000 invitation codes for dear browsers to skip the wait list, but only for AI profit boardroom members. If you want to book a free SEO strategy session with my team, the link is in the comments and description. We'll show you exactly how to use AI and SEO to get more customers. And if you want access to our complete AI automation system with over 100 use cases, check out the AI success lab. Link is in the comments and description. You'll get step-by-step tutorials, templates, and processes for everything we talked about today, plus way more advanced AI strategies that most people don't even know exist. Look, AI browsers are just the beginning. The next few years are going to bring AI tools that make today's technology look like stone age tools. The question is, are you going to be ready or left behind while everyone else gets ahead? The choice is yours. But don't wait. Technology moves fast. The people who adapt quickly win. The people who wait lose. Start with these AI browsers today. Join the AI profit boardroom. Book that strategy session. Get into the AI success lab. Your future self will thank you. Julian Goldie reads every comment, so drop your thoughts below. Which AI browser are you going to try first? And what other AI tools do you want me to test next? See you in the next video. Norton Neo new AI browser is absolutely insane. Today, I'm going to show you the most insane AI browser I've ever seen. This thing just changed everything about how we browse the web. It's called Norton Neo, and it's not just another browser with AI stuck on top. This was built from the ground up with AI at its core. The stuff it can do will blow your mind. And the best part is completely free to download right now. Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency Goldie Agency. Whilst he's helping clients get more leads and customers, I'm here to help you get the latest AI updates. Julian Goldie reads every comment, so make sure you comment below. Look, I've been using browsers for over 20 years. Chrome, Firefox, Safari, Edge, they're all basically the same thing with a fresh coat of paint. But Norton Neo, this is different. This is what browsers should have been from day one if AI existed back then. I'm going to show you exactly what makes this so special. And trust me, by the end of this video, you're going to want
Segment 27 (130:00 - 135:00)
to download this immediately. But first, let me tell you what's wrong with every browser you're using right now. You open Chrome, you search for something, you click a link, you open another tab, then another. Before you know it, you have 47 tabs open and you can't find anything. Sound familiar? Then you're switching between apps, copying and pasting, trying to organize all this mess. It's chaos. We're browsing the web like we did 20 years ago, searching, clicking, switching tabs, drowning in content and distractions. It's 2025 and we're still doing things like cavemen. Norton Neo fixes all of this and I'm going to show you exactly how. First thing you need to know about Neo is something called the magic box. This is your AI command center. Think of it like having a super smart assistant that never leaves your side. Instead of opening new tabs and switching between apps, you just type what you want in the magic box. I tested this myself. I asked it to find the three most important quantum computing developments from the past month. Within seconds, it gave me three perfect summaries with sources and publication dates. No clicking around, no opening 20 different websites, just instant, accurate answers with proof. But here's where it gets crazy. I also asked it to find the five most cited Crisper research papers from 2024. Again, perfect results. Title, authors, year, and a summary for each one. This would have taken me hours to research manually. Neo did it in seconds. The magic box isn't just search. It's search, chat, and action all rolled into one. You can ask it to summarize, generate content, research topics, compare products, anything. It's like having chat GPT, Google, and a research assistant all in one place. Now, let me show you something that will change how you browse forever, and it's called peak and summary. You know how annoying it is when you see a link, but you're not sure if it's worth clicking. With Neo, you don't have to click. You can preview any website right in the browser. I tested this when looking for Italian pasta recipes. Instead of clicking link after link, I could preview each website, scroll through the content, and even interact with it without actually visiting the page. I found exactly what I needed in half the time. But wait, it gets better. I tried this with PDF conversion tools. I could preview the websites, see their features, and even upload files to test them without ever leaving Neo. This is insane. You're literally browsing the web without browsing the web. And here's the key. Neo automatically categorizes your tabs. Email tabs go in one section, AI chats in another, research in another. No more hunting through dozens of random tabs. Everything is organized automatically. But the real game changer is how Neo eliminates tab chaos completely. Instead of tabs, Neo uses something called sessions. Think of sessions like smart folders for your browsing. When you're researching a topic, all related pages get grouped together. When you're planning a trip, that gets its own session. When you're working on a project, that gets its own session, too. You can pick up exactly where you left off on any session. No more losing your place. No more trying to remember which tab had that important information. Neo remembers everything for you. Now, let's talk about the AI typing assistant. This thing makes every keystroke smarter. Whether you're writing emails, social media posts, or documents, Neo's AI helps you write like a pro, even if you're not one. Type less, say more. It suggests better words, fixes your grammar, and makes your writing clearer. It's like having a professional editor built right into your browser. But here's what really sets Norton Neo apart from every other browser. Security. This browser is equipped with Norton Web Shield. That means it blocks fishing, malware, and unsafe sites in real time. Your browsing stays clean and secure automatically. It has built-in ad blocking, too. No more annoying pop-ups. No more slow loading pages because of ads. Just clean, fast browsing with trusted Norton protection. Here's something that will blow your mind. Neo learns what matters to you. Not in a creepy way, but in a helpful way. If you're researching AI tools, it starts showing you relevant content. If you're planning a vacation, it surfaces travel deals and guides. If you're learning a new skill, it finds related tutorials and resources. It creates a personalized content feed just for you. But unlike social media feeds that waste your time, Neo's feed actually helps you get things done. And Neo remembers everything, it proactively surfaces helpful reminders and content based on your past activity, so nothing slips through the cracks. If you were researching something last week and new information becomes available, Neo will let you know. Let me give you a real example. Say you're researching the best AI writing tools. You compare Grammarly, Jasper, and Write Sonic. Neo will remember this. Later, when one of these tools updates their pricing or features, Neo will proactively tell you about it. You don't have to remember to check back. Neo does the remembering for you. This isn't just helping you browse. It's helping you think, learn, create, and focus. It's like having a research assistant that never sleeps and never forgets. I also tested Neo's fact-checking abilities. I asked it to verify the claim that humans only use 10% of their brains. Within seconds, it
Segment 28 (135:00 - 140:00)
found three credible scientific sources that debunked this myth. It summarized each source's conclusion and told me which stance each one took. This kind of research would normally take me 30 minutes. Neo did it in 30 seconds. The speed is incredible. Every search is fast. Every response is accurate. Every source is credible. I've been testing this for weeks and I haven't found a single wrong answer yet. But let me tell you about my favorite feature, the preview system I mentioned earlier. This changes everything about how you research online. Instead of opening dozens of tabs and getting lost in information overload, you can preview content instantly. I was researching how to convert PDFs to Word documents. Normally, I'd click on 10 different websites, compare their features, try their tools, and waste an hour figuring out which one actually works. With Neo, I previewed each site, tested their tools right in the preview, and found the best solution in 5 minutes. You can even upload files to websites through the preview. This means you can test tools and services without actually committing to visiting the full website. It's like window shopping but for the entire internet. Neo also categorizes everything automatically. When you're browsing for business, those tabs get grouped together. When you're shopping, that gets its own category. When you're researching, that gets sorted separately. Your browser stays organized without you having to do anything. The sessions feature is genius, too. Let's say you're planning a marketing campaign. All your research, competitor analysis, and planning documents get grouped into one session. You can come back to it days later and everything is exactly where you left it. Or maybe you're learning about AI automation. Neo creates a session for that and groups all your tutorials, tools, and resources together. No more bookmarking random pages and forgetting what they were for. This is proactive browsing. Neo isn't just reacting to what you do. It's actively helping you get better results. It's predicting what you need before you need it. It's organizing your work before you even know it needs to be organized. And because it's powered by Norton, you know your data is safe. Your browsing history isn't being sold to advertisers. Your personal information isn't being harvested. Norton has been protecting people online for decades. They know what they're doing. The best part, Neo is completely free. You can download it right now and start using all these features immediately. No subscription, no premium tiers. Just download and go. I've been using Neo for weeks now and I can't go back to regular browsers. They feel slow and dumb compared to this. It's like going from a smartphone back to a flip phone. Technically, it still works, but why would you want to? Neo is smarter, cleaner, more secure, and more proactive than any browser I've ever used. It's not just a browser with AI features bolted on. It's what Chrome would be if it were built today with AI at the core. This is the future of web browsing, and the future is available right now. Neo isn't just changing how we browse. It's changing how we work, research, and think online. Download it today and see for yourself why this is the most insane AI browser ever created. The link is in the description, and make sure you join their Discord community, too. The link is also in the description. What do you think about Neo? Are you going to try it? Let me know in the comments below. Julian reads every single comment, so make sure you share your thoughts. And if this video helped you understand why AI browsers are the future, smash that like button and subscribe for more AI updates. I'll see you in the next one. Today, I'm going to show you Perplexity's brand new comet tool that just dropped. This thing is absolutely insane. It's like having a team of researchers working for you 24/7. We're talking about deep research that would take you hours in just minutes. This is going to change everything about how you research online. And the best part is so easy my grandma could use it. Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency Goldie Agency. Whilst he's helping clients get more leads and customers, I'm here to help you get the latest AI updates. Julian Goldie reads every comment, so make sure you comment below. So, here's what just happened. Perplexity just launched their brand new browser called Comet. And this isn't just another browser. This is the first AI browser that actually thinks for you. While everyone else is trying to stuff AI into Chrome, Perplexity said, "Forget that. We're building our own browser from scratch. " And the timing is perfect. Google just got hit with antitrust stuff. Chrome might get broken up. And here comes Perplexity with the perfect replacement. But here's the crazy part. This browser doesn't just search. It actually does the work for you. Let me explain what I mean. You know how you normally search for something on Google, you type in your question, you get a bunch of links, you click through 10 different websites, you read through tons of fluff, you try to piece together the answer yourself, it takes forever. Comet flips all of that upside down. Instead of you doing all that work, the AI does it for you. You ask a question. Comet goes out and reads dozens of websites. It pulls out all the important stuff. It puts together a complete answer with sources all in about 2 minutes. They're calling this a gentic search, which is a fancy way of saying that AI acts like your personal research assistant. It doesn't just find information, understands what you need, it makes decisions about what's important. It even follows up on things you didn't think to ask about. Here's
Segment 29 (140:00 - 145:00)
how powerful this is. Let's say you want to research best marketing strategies for local businesses. With normal Google search, you'd spend hours clicking through blog posts and marketing sites. Half of them would be trying to sell you something. the other half would have outdated information. With Comet, you ask that same question. The AI goes out and reads through hundreds of sources. It finds the most recent studies. It looks at what's actually working right now. It pulls out the key strategies. It even gives you specific steps to implement them all in one clean report with sources you can verify. But here's where it gets really crazy. Comet remembers everything you've looked at. So, if you were researching local marketing last week and today you want to know about social media ads, it connects those dots. It might say, "Based on your interest in local marketing, here are the social media strategies that work best for local businesses. " This contextual awareness is what makes Comet different from everything else out there. It's not just answering your current question. It's building on everything you've been working on. It's like having a research partner who never forgets and never gets tired. And the browser itself is built on Chromium, so all your Chrome extensions work. All your bookmarks transfer over. You're not losing anything. you're just gaining this massive AI upgrade. Now, let me show you some of the specific features that make this so powerful. First is the Comet Assistant. This loss right in your browser. You can ask it to do things like summarize all your open tabs, close duplicate tabs, find that article you were reading about SEO last Tuesday, schedule meetings based on emails you've received. Think about how much time you waste managing your browser. You got 20 tabs open. You can't remember which one had that important information. You're clicking back and forth trying to find things. Comet Assistant handles all of that for you. Here's a real example. You're researching a new business idea. You've got tabs open for market research, competitor analysis, pricing strategies, and legal requirements. Instead of trying to keep track of all this in your head, you just ask Comet to summarize all my research on this business idea. The AI looks at everything you've been reading. It pulls out the key points from each area. It creates a comprehensive summary that connects all the pieces. It might even point out things you missed or suggest additional research areas. The deep research feature is where this gets really impressive. You can ask Comet to do a deep dive on any topic. It spends several minutes reading through dozens of sources. It's not just skimming. It's actually analyzing the content, looking for patterns, checking facts across multiple sources. When it's done, you get a report that would normally take a human researcher hours or days to complete. And it's not just a summary of what other people wrote. It's an analysis that connects ideas from different sources and gives you insights you wouldn't have found on your own. Based on what we're seeing from early users and demos, people are getting incredible results with business research. Users report asking it to analyze markets and getting back insights about market size, growth trends, key players, and emerging opportunities. The reports include data from recent studies that would normally take hours to find manually. The scary part, this level of research used to be something only big companies could afford. They'd hire teams of analysts to put together reports like this. Now, anyone can get the same quality research in minutes instead of weeks. But here's what really sets Comet apart from other AI browsers that are coming out. Most of them are just adding AI features to existing browsers. Comet was built from the ground up to be AI native. Every part of the browser is designed to work with AI. The interface adapts to how you work. If you're doing research, it surfaces relevant tools and information. If you're writing, it offers different assistance. If you're shopping, it helps you compare products and find the best deals. It's not just one AI feature stuck into a regular browser. It's a completely AI powered browsing experience. And they've thought about privacy in a way that most companies haven't. You can control exactly what data Comet uses. You can opt out of data sharing completely if you want. It has built-in ad blocking and tracker protection. So, you're getting the AI benefits without sacrificing your privacy. This is huge because privacy has been the main concern people have about AI browsers. Nobody wants their browsing data getting sold to advertisers or used to train AI models without permission. Comet gives you control over all of that. Now, let's talk about who this is for. Obviously, anyone who does research for work is going to love this. That includes marketers, writers, students, consultants, lawyers, doctors. Basically, anyone whose job involves finding and analyzing information. But it's also perfect for personal use. Planning a vacation? Comet can research destinations, compare prices, find the best activities, and create a complete itinerary. Shopping for a new car. It can analyze reviews, compare features, find the best deals, and even help you negotiate with dealers. The key is that Comet doesn't just find information. It processes it for you. It thinks about what you actually need to know. It saves you from having to read through tons of irrelevant content to find the good stuff. Actually, I just got access to Perplexity Max, so I'm going to be testing Comet Live for you guys. I'm
Segment 30 (145:00 - 150:00)
planning to run some real business research tests and show you exactly what this thing can do. At $200 a month, it sounds expensive until you think about what you're getting. How much would you pay for, you know, a personal research assistant who works 24/7 and never makes mistakes? Probably a lot more than $200 a month. But here's the thing. Perplexity has said they're going to roll this out more broadly. So, even if you can't afford the Max plan right now, you'll probably be able to access Comet soon at a lower price point. And honestly, if you're running a business, $200 a month for this level of research capability could be a gamecher. Think about how much time your team spends on research. Think about how much better your decisions would be if you had access to comprehensive, accurate information instantly. Based on early user reports, this tool is already saving people dozens of hours on content research, competitor analysis, and market research. The insights they're getting are reportedly way better than what they could find manually. I'm going to be testing this myself and documenting the results. I want to see if it really can replace hours of manual research with minutes of AI powered analysis. Stay tuned for that case study. This focus makes the experience so much better. When you open Comet, you're not distracted by news feeds or notifications. You're in work mode. You can focus on what you're trying to accomplish without getting pulled in different directions. The company behind this is also impressive. Perplexity has raised over $500 million. They're valued at 9 billion. They're processing over a 100 million queries per week. They're not some startup that might disappear next year. They have the resources to compete with Google long-term. And they're hiring talent from Google's Chrome team. So they understand how browsers work at a deep level. They're not just building AI features. They're building a better browser that happens to have AI built in. Looking at the bigger picture, this represents a massive shift in how we interact with information online. Instead of us adapting to how computers want to organize information, computers are finally adapting to how humans actually think and work. For years, we've had to learn Google's language. We've had to figure out the right keywords. We've had to know which sites have good information. We've had to develop skills for filtering through search results. Comet flips all of that. You can ask questions in plain English. You can have conversations about what you're trying to learn. You can get answers that actually help you make decisions instead of just giving you more information to sort through. This is especially powerful for complex topics where you need to understand multiple perspectives or connect information from different sources. Traditional search is terrible at this. You end up with disconnected pieces of information that you have to somehow put together yourself. Comet excels at connecting dots across different sources. It can identify patterns and relationships that you might miss. It can point out contradictions in different sources. It can help you understand the credibility and bias of different information sources. For business owners, this changes everything about competitive intelligence and market research. Instead of spending weeks trying to understand your market, you can get comprehensive insights in a day. Instead of wondering what your competitors are doing, you can get detailed analysis of their strategies and performance. For content creators, it's like having a research team at your fingertips. You can quickly dive deep into any topic. You can find unique angles that others haven't covered. You can back up your content with solid research instead of just opinions. For students and academics, it's a gamecher for thesis research and literature reviews. Instead of spending months reading through papers to understand the current state of research in your field, you can get a comprehensive overview in hours. The applications are endless. And we're just seeing the beginning. As more people start using Comet, the AI is going to get better. It's going to understand more nuanced requests. It's going to develop better insights about what information is actually useful versus what's just noise. One thing that's really interesting is how this might change content creation on the web. When AI can quickly identify the most useful and accurate information, there's going to be huge pressure on websites to create genuinely valuable content instead of just SEO optimized fluff. This could actually improve the quality of information online. When algorithms reward content that's actually helpful rather than content that just ranks well, creators will start focusing on providing real value. But here's something most people aren't thinking about. Comet isn't just a research tool. It's a learning accelerator. When you can quickly get comprehensive information about any topic, you can learn much faster. You can explore ideas you never would have had time to research before. I'm going to be testing this with some real business scenarios and document everything. I want to see if the hype matches reality when it comes to actual business research that impacts the bottom line. I'm even going to test it on my own YouTube channel. Have comment analyze and summarize my latest videos to see what insights it can pull out about my content strategy. This democratizes expertise in a way we've never seen. Skills and knowledge that used to take years to develop can now be acquired much faster. Not completely. You still need experience and practice, but the information gathering and analysis part which used to be a huge barrier is now almost effortless. Think about what this means for entrepreneurs. How many great
Segment 31 (150:00 - 155:00)
business ideas never get started because the research phase seems too overwhelming? How many people give up on learning new skills because the information gathering process is too timeconuming? Comet removes those barriers. It makes research so fast and easy that you can afford to explore ideas you might not have pursued before. You can quickly validate or invalidate assumptions. You can understand markets and opportunities much faster. The productivity gains are incredible. But the real power is in the exploration it enables. When research isn't a bottleneck, you can follow your curiosity wherever it leads. You can ask what-if questions and actually get answers quickly enough to matter. For agencies and consulting businesses, this is transformative. You can provide much deeper analysis to clients without increasing your costs. You can take on more complex projects because the research component doesn't require huge time investments. You can also differentiate yourself by providing insights that competitors can't match. When you have access to comprehensive AI powered research capabilities, you can offer value that others simply can't provide with manual research methods. Now, let me be clear about something. Comet isn't perfect. No AI tool is. You still need to think critically about the information you get. You still need to verify important facts. You still need to use your judgment about what information is relevant to your specific situation. But it's incredibly good at the heavy lifting of research. It can process far more information than any human could in the same amount of time. It can identify patterns and connections that would be easy to miss. It can synthesize information from multiple sources in ways that are genuinely helpful. The key is understanding what comet is good at and what it's not. It's excellent at finding information, analyzing trends, comparing options, and creating comprehensive overviews. It's not perfect at creative thinking, strategic decision-m or understanding highly nuanced contexts that require deep industry experience. But when you combine Comet's research capabilities with human creativity and judgment, the results are powerful. You get the best of both worlds. Comprehensive information gathering plus human insight and decision-m. Looking forward, I think we're going to see more tools like this. Google is already adding AI features to Chrome. Microsoft is upgrading Edge with AI capabilities, but Perplexity has a head start because they built Comet from the ground up to be AI native. They also have an advantage because search is their core business. They understand information retrieval and analysis better than companies that are primarily focused on other things. This focus shows in the quality of comt research capabilities. The browser wars are about to get very interesting. For years, Chrome has dominated because it was fast and had good developer tools. But speed and developer tools aren't enough anymore. Users want intelligence. They want browsers that help them get things done instead of just displaying web pages. If you want to stay ahead of the curve and get access to cuttingedge AI tools like Comet, you need to join my AI profit boardroom. This is the best place to scale your business, get more customers, and save hundreds of hours with AI automation. We're constantly testing and sharing the latest AI tools that can give you a competitive edge. And if you're serious about using AI to grow your business, book a free SEO strategy session with my team. The link is in the comments and description. We'll show you exactly how AI tools like Comet can be integrated into your marketing strategy to get better results with less effort. Finally, if you want step-by-step training on implementing AI in your business, check out our AI success lab. We have SOPs and processes for over a 100 different AI use cases. You'll get access to video tutorials, templates, and a community of 14,000 members who are all implementing AI in their businesses. This is just the beginning of the AI revolution in how we access and use information online. The companies and individuals who adapt quickly are going to have massive advantages over those who wait. Don't be the person who wishes they had started using these tools sooner. Make sure you're subscribed and hit the bell because I'll be sharing those results with you. Today, I'm going to show you the most insane AI browser that just dropped. This thing actually does your work for you. It goes to websites, buys stuff, sends emails, and books meetings while you sit back and watch. I tested it with 10 crazy tasks, and the results will blow your mind. This is not just another AI tool. This is the future of how we use the internet. Get ready because everything is about to change. Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency Goldie Agency. Whilst he's helping clients get more leads and customers, I'm here to help you get the latest AI updates. So, Perplexity just released something called Comet. And this isn't just another chatbot. This is an AI that can actually control your browser. It can click buttons, fill out forms, buy products, send emails, and do pretty much anything you can do online. But instead of you doing it, the AI does it for you. I spent hours testing this thing with 10 different tasks. Some worked perfectly, some didn't work at all, and some results shocked me. so much I had to test them twice. By the end of this video, you'll know exactly what this AI can do, what it can't do, and whether you should be using it right now. But first, let me explain what
Segment 32 (155:00 - 160:00)
Comet actually is. Most AI tools just give you text answers. You ask a question, they give you an answer. That's it. Comet is different. It can actually browse the web like a human. It can go to Amazon, compare prices, add items to your cart, and even check out. It can go to your email, read messages, and reply to them. It can schedule meetings, book restaurants, and do research across multiple websites. Think about how much time you spend doing boring tasks online, checking prices, sending emails, looking up information, scheduling stuff. What if an a I could do all of that for you? That's exactly what Comet promises to do. Now, before I show you my tests, you need to understand how this works. When you give Comet a task, it doesn't just search the web like Google. It actually opens websites in a browser. It can see the page just like you can. It can read text, click buttons, fill out forms, and navigate between pages. It's like having a virtual assistant that can use any website. But here's the crazy part. It can do multiple tasks at once. While it's checking prices on Amazon, it can also be looking up reviews on another site. While it's composing an email, it can be researching information to put in that email. This isn't just automation. This is like having multiple people working for you at the same time. All right, let's get into my tests. I gave Comet 10 different tasks to see what it could actually do. Some of these tasks were simple, some were complex, and some were designed to break it. Test number one was cleaning up my email. I told it to unsubscribe me from all recent marketing emails in my inbox, especially from brands like Grock X and Google. Now, this is something that takes me forever to do manually. You have to open each email, scroll to the bottom, click unsubscribe, confirm you want to unsubscribe, and then do it again for the next email. Comment went straight to my Gmail. It opened up the marketing emails. It clicked the unsubscribe links. It confirmed the unsubscriptions. And it did this for multiple emails automatically. In about 2 minutes, it had cleaned up my inbox better than I could have done in 20 minutes. This alone saved me so much time and frustration. Test number two was scheduling a dinner with friends. I told it to schedule a dinner with Sarah, Mike, and James for Friday at 7:00 p. m. at a restaurant we all like in central London and to create a group message to confirm. This is a complex task because it involves multiple steps and coordination. Here's what blew my mind. Comet didn't just send one email. It created a proper group email with all three people. It suggested restaurant options in central London. It created a Google Meet meeting as backup. And it even included a calendar invite. This is the kind of task that usually takes me 15 minutes of back and forth emails. Comet did it all in one go. Test number three was price comparison shopping. I asked it to compare the price of an Instant Pot six quart on Walmart and Target to see which one was cheaper, including shipping. This is something most people do manually by opening multiple tabs and checking prices. Comet went to both websites. It searched for the exact product. It found the prices. It checked shipping costs and it gave me a clear comparison with links to both products. It even told me which one was the better deal. This kind of research usually takes 10 to 15 minutes. Comet did it in under 3 minutes. Now, here's where I want to pause and talk about something important. If you're running a business or trying to scale your operations, these kinds of time savings add up fast. Imagine having an AI that can do your research, handle your emails, and manage your online tasks while you focus on the big picture stuff. That's exactly what we teach inside the AI success lab. We show you how to use AI tools like this to save hundreds of hours and grow your business faster. We have over 14,000 members who are already using AI to automate their work and increase their profits. The link is in the comments and description, and you'll get access to over 100 AI use cases, step-by-step tutorials, and all the video notes from trainings like this one. All right, back to the tests. Test number four was finding a recipe. I asked Comet to find a quick and healthy recipe for chicken stir fry that takes under 30 minutes to cook. This seems simple, but I wanted to see how good it was at finding specific information. Comet found multiple recipe options. It filtered them based on my criteria, quick, healthy, and under 30 minutes. It gave me ingredient lists, cooking instructions, and even nutritional information. What impressed me was how it compared different recipes and recommended the best one based on my requirements. Test number five was general browsing and research. I told it to browse the web to find the top three trending books on productivity right now and where to buy them cheapest. This requires research across multiple websites and price comparison. Comet found three books. Atomic Habits by James Clear, Deep Work by Cal Newport, and Getting Things Done by David Allen. But it didn't stop there. It checked prices across Amazon, Target, and other retailers. It showed me where each book was cheapest. It even included links so I could buy them immediately. This is the kind of research that would take me at least 30 minutes. Comet did it perfectly in under 5 minutes. Test number six was the most impressive one. I asked it to find a recipe for vegan lasagna, then add all the ingredients needed to my Walmart cart automatically.
Segment 33 (160:00 - 165:00)
This combines research with actual online shopping and cart management. Comet found a vegan lasagna recipe. It made a complete ingredient list. Then it went to Walmart's website, searched for each ingredient, and added them to my cart. I'm talking about specific items like noya refrigerated soy extra firm organic tofu, Great Value smoked paprika, Great Value garlic powder, and everything else needed for the recipe. This blew my mind because it's not just finding information. It's actually taking action on that information. It's making purchases. It's managing my shopping cart. This is like having a personal assistant who can actually buy stuff for you online. Test number seven was YouTube research. I asked Comet to go to Julian Goldieco's YouTube channel, that's my channel, and summarize the main tips from my latest video. I wanted to see if it could analyze video content accurately. Comet went to my YouTube channel. It found my latest video about lovable. It watched or analyzed the content and gave me an accurate summary of the main points I covered in that video. It even pulled out specific tips and actionable advice. This shows it can understand and analyze video content, not just text. Test number eight was researching chat GPT5 leaks. I told it to find all the latest info about chat GPT5 leaks, including possible release date, features, and any verified updates from OpenAI. This is current information that changes daily. Comet searched across multiple websites and news sources. It found the latest information about GPT5. It even mentioned that Sam Alman discussed it in a recent podcast and that it might be released in August. The research was current, accurate, and comprehensive. This is the kind of research that would take hours if you did it manually. Test number nine was another podcast research task. I asked it to summarize Sam Alman's most recent podcast interview and pull out key takeaways, quotes, and new AI insights. This requires finding the right podcast, analyzing the content, and extracting the most important information. Comet found the most recent podcast with Sam Alman. It gave me a detailed summary with specific quotes. It highlighted the key insights about AI development. It even organized the information in a way that was easy to understand and actionable. This is like having a research assistant who can analyze hours of content in minutes. Julian Goldie reads every comment, so make sure you comment below and let me know what you think about these results. Test number 10 was the most complex one. I asked it to find upcoming AI conferences or webinars happening this month and add the top two most relevant ones to my Google calendar with links and times. This combines research with calendar management and requires integration with external tools. Comet search for AI events happening in the current month. It evaluated which ones were most relevant based on my background and interests. Then it actually added events to my Google calendar with all the details including links, times, and descriptions. This is end to end automation that saves hours of manual work. Now, let's talk about what this all means. Out of 10 tests, Comet performed exceptionally well on eight of them. The two that didn't work perfectly were still impressive attempts. This isn't just a research tool or a chatbot. This is a digital assistant that can actually take actions on your behalf. But here's what most people are missing. This technology is still new. Most people don't even know it exists yet. The people who figure out how to use tools like this first are going to have a massive advantage. While everyone else is manually doing research, sending emails, and managing tasks, you'll have an AI doing it all for you. Think about your daily routine. How much time do you spend on repetitive online tasks, checking prices, researching information, sending emails, scheduling meetings, managing your calendar? What if you could get two to three hours back every day by having an a I handle these tasks? That's not just time savings. That's business transformation. You could focus on strategy instead of busy work. You could work on growth instead of maintenance. You could spend time with family instead of being stuck at your computer doing boring tasks. But there's a catch. Tools like Comet aren't just plugandplay. You need to know how to use them effectively. You need to understand what they can and can't do. You need to know how to write good prompts that get the results you want. This is exactly why I created the AI Profit Boardroom. It's where I teach business owners how to use AI tools like this to scale their operations and save hundreds of hours every month. We currently have over 1,000 members who are using AI to automate their businesses and increase their profits dramatically. Inside the AI profit boardroom, you learn how to identify which tasks can be automated, how to set up AI systems that work reliably, and how to scale your business while working fewer hours. The best part is you get direct access to me and a community of other entrepreneurs who are all using AI to grow their businesses. But remember, AI tools are only as good as the person using them. You need to understand how to write effective prompts. You need to know how to break down complex tasks into steps the AI can handle. You need to know how to check and verify the AI's work. This is why training and education are so important. You can't just buy an AI tool and expect it to magically solve all your problems. You need to learn how to use it effectively. You need to understand its capabilities and
Segment 34 (165:00 - 170:00)
limitations. You need to develop systems and processes that incorporate AI into your workflow. If you want to get serious about using AI to grow your business and save time, I strongly recommend checking out the AI success lab. We have step-by-step tutorials for over 100 different AI use cases. We show you exactly how to set up automation systems. We give you the prompts and templates you need to get started immediately. Plus, you get access to a community of 14,000 other entrepreneurs who are all using AI to scale their businesses. You can see what's working for them, ask questions, and get help when you're stuck. The link is in the comments and description below. Now, if you're running an SEO agency or any kind of service business, I also want to mention that we're offering free SEO strategy sessions. These are one-on-one calls where we analyze your current SEO situation and show you how to get more leads and customers. The link for that is also in the comments and description. Let me wrap up with the key takeaways from testing Plexity comment. First, AI browsers are real and they work. This isn't science fiction anymore. You can have an AI that actually browses the web and takes actions for you right now. Second, the time savings are significant. Tasks that take 20 to 30 minutes manually can be done in two to three minutes with AI. If you're doing multiple tasks like this every day, that adds up to hours of time saved. Third, this technology is still early but improving rapidly. The people who learn to use it now will have a big advantage as it gets better and more widespread. Fourth, you need proper training to use these tools effectively. They're powerful, but they require skill to use properly. You can't just wing it and expect good results. Finally, this is just the beginning. AI browsers like Comet are going to transform how we work online. The question isn't whether this technology will become mainstream. The question is whether you'll be ready when it does. So, start experimenting now, start learning now, start building AI into your workflow now. Cuz the businesses that figure this out first are going to dominate their markets while everyone else is still doing things the old way. And remember, if you want help implementing AI in your business, check out the AI Profit Boardroom where we show you exactly how to use tools like this to scale your operations and increase your profits, we have over 1,000 members who are already seeing incredible results. Thanks for watching and I'll see you in the next video where we'll dive deep into more AI tools that are changing everything. Today, I'm going to show you which AI browser actually works and which ones are total garbage. I tested all three of the biggest AI browsers right now. One of them failed so bad it couldn't even answer basic questions. Another one was so slow I thought my internet broke. But one browser shocked me with how good it actually was. You're about to see the real results that nobody else is showing you. Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency Goldie Agency. Whilst he's helping clients get more leads and customers, I'm here to help you get the latest AI updates. So AI browsers are everywhere now. Everyone's talking about them. But here's the thing. Nobody tells you. Most of them suck. Like really suck. I spent hours testing three of the biggest AI browsers out there. Perplexity, Comet, Genens, Spark, and Dia. And what I found will shock you. One of these browsers couldn't even handle basic questions without crashing. Another one was so slow I could make a sandwich while waiting for answers. But one browser actually blew my mind with how good it was. and I'm going to show you exactly what happened when I put them through six brutal tests. But first, let me tell you why this matters to you. If you're using the wrong AI browser, you're wasting hours every single day. You're getting bad information. You're making wrong decisions and you're falling behind everyone else who figured this out already. The right AI browser can save you 20 hours a week. It can give you better research in minutes instead of hours. And it can help you make money faster than ever before. So, let's dive into what I found. What? Test one, search accuracy. The first test was simple. I asked each browser one basic question. Who won the Nobel Prize in physics in 2023? This should be easy, right? Wrong. Genpark immediately gave me a request aborted error. It couldn't even try to answer. I had to ask again just to get a response. That's already a red flag. Dia gave me a short answer. It was correct, but barely any details like getting a text from your friend who doesn't want to talk. But perplexity, it gave me the exact answer with details and context. It told me who won and why they won. This is what I want from an AI browser. But wait, it gets worse for the other browsers. Test two, real time information. Next, I tested something harder. I asked for Tesla's current stock price. This needs real time data. It changes every minute. And here's where things got really ugly. Genpark failed again. Same request aborted error. Even when I tried again, it barely worked. How can you trust a browser that crashes half the time? The gave me a basic answer. Just the price, no chart, no context, nothing useful, but perplexity. It gave me everything. Current price, detailed stats, graphs, the works. It was like having a
Segment 35 (170:00 - 175:00)
professional financial analyst right there. And this is just the beginning. Wait until you see what happened with the speed test. Test three, user experience. I wanted to see how these browsers actually feel to use. So, I searched for best Italian restaurants in New York. Simple search, right? But the results were crazy different. Genpark was the slowest. It gave me some restaurant names, but no links, no images, no way to actually find these places. What's the point? I did better. It gave me images and links, but the details were weak. Like someone who knows restaurants exist, but never actually eaten at them. Perplexity crushed it. Images, ratings, sources, everything you need. It was like having a local food expert as your best friend. But the speed test is where one browser completely embarrassed itself. Test for speed. I asked the simplest question possible like what's the capital of Australia? This should take one second, maybe two. Dear and Perplexity both answered fast, like instantly fast. But Genspark, I waited and waited some more. It was so slow I thought my computer broke. for a basic question that a 5-year-old could answer. If you can't answer basic questions quickly, how can I trust you with complex research? But the next test is where I found out which browsers you can actually trust. Key test five, source verification. I asked a specific question. According to the CDC, how many people contracted influenza last year? This test if the browsers actually check their sources because making stuff up is easy. Proving it is hard. And guess what happened? GenSpark failed again. Request aborted. At this point, I was getting angry. How is this browser even still running? Dear didn't respond at all, just ignored the question. Like pretending you didn't hear someone when they ask you to do work. Only perplexity gave me a real answer with actual CDC sources with links I could check. This is what professional research looks like. But the privacy test revealed something that might change your mind about all of them. Test six, privacy. I tested how these browsers handle tracking and cookies. Because your data matters, I visited CNBC, which is loaded with trackers. Then I checked what each browser did. Perplexity analyzed the website and told me exactly what was happening. Full transparency. Dear and Genspark gave generic answers. Nothing specific. No real protection details. This matters because your privacy is worth money. Companies pay billions for your data. If your browser isn't protecting you, you're getting robbed every day. After all these tests, here's what I found. Perplexity Comet won every single category. Search accuracy, real-time data, user experience, speed, source verification, and privacy analysis. It wasn't even close. Genpark failed so many tests, I couldn't believe it. Multiple request aborted errors, slow responses, missing features. It's like a car that only works on Tuesdays. Deer was okay, but basic. It did some things right, but never impressed me. It's like that friend who shows up but doesn't really help. But here's the thing. Nobody tells you about AI browsers. Most people pick the wrong one and then complain that AI doesn't work. They blame the technology when they should blame their choice. It's like buying a broken calculator and then saying math is stupid. The right AI browser can change everything for you. Better research, faster answers, more accurate information, and that leads to better decisions, more money, and less stress. But the wrong AI browser wastes your time and gives you garbage results. So which one should you actually use? Based on my testing, Perplexity Comet is the clear winner. It worked when others failed. It was fast when others were slow. It gave details when others gave nothing. But here's what's really crazy. Most people will watch this video and then do nothing. They'll keep using whatever browser they're using now. Even if it's slow, even if it's wrong, even if it's wasting their time. Don't be that person. If you want to save 20 hours a week and get better results, switch to the browser that actually works. But picking the right browser is just the beginning. You also need to know how to use AI tools to actually make money. How to automate your business. How to get more customers with less work. That's where my AI profit boardroom comes in. It's the best place to scale your business, get more customers, and save hundreds of hours with AI automation. We currently have over 1,000 members who are crushing it with AI. These are business owners who figured out how to use AI to 10x their results. They're making more money, working less hours, and beating their competition. If you want to join them, the link is in the comments and description. But maybe you're thinking, "Julian, this sounds great, but I need help with my specific business. " I get it. Every business is different. That's why I offer free SEO strategy sessions where we look at your exact situation and create a plan to get you more traffic and customers. These sessions are completely free. No pitch, no pressure, just value. The link to book is in the comments and description. And if you want to go even deeper, check out
Segment 36 (175:00 - 180:00)
my AI success lab. We have an SOP and process for everything. Over 100 different use cases, video tutorials, step-by-step guides. You see, I give away tons of free content every single day. Checklists, tutorials, insider tips, all inside the school feed. You can get all the video notes from this content and everything else, plus all the trainings in the AI success lab. Right now, we have over 14,000 members. That's 14,000 people who decided to stop making excuses and start getting results. They're part of something bigger than themselves. A community of winners who help each other succeed. The link is in the comments and description. And but here's the thing about AI browsers and AI tools in general. The technology is moving so fast that what works today might not work tomorrow. New browsers are launching every month. New features are coming out every week. That's why you need to stay updated. test things yourself. And that's why you need a community of people who are doing the same thing. Because the people who win with AI aren't the ones who find one tool and stick with it forever. They're the ones who keep testing, keep learning, and keep adapting. They're the ones who understand that AI is not just about tools. It's about systems. It's about processes. It's about thinking differently. And that's exactly what we teach inside the AI success lab. Look, I've tested hundreds of AI tools. I've built a sevenf figureure agency using AI and automation. I've helped thousands of students get results with AI. And the biggest mistake I see people make is thinking that one tool will solve all their problems. It won't. Success with AI comes from using the right combination of tools in the right way for your specific situation. That's what separates the winners from everyone else. The winners have systems. They have processes. They have communities. They have support. The losers have shiny object syndrome. They jump from tool to tool hoping for magic. Which one are you going to be? Julian Goldie reads every comment. So, make sure you comment below and let me know which AI browser you're going to try. Are you going with Perplexity Comet based on my testing or are you going to stick with what you're using now? And more importantly, what are you going to do with the time you save? Because that's the real question. AI browsers and AI tools don't matter unless they help you achieve your goals faster. Whether that's growing your business, making more money, or just having more time for the things you love, the choice is yours. But whatever you choose, choose fast. Because while you're thinking about it, your competition is already using these tools to get ahead. And in business, being second is the same as being last. So go test perplexity comet. Book that free strategy session. Join the AI success lab. But most importantly, take action because knowledge without action is just entertainment. And I'm not here to entertain you. I'm here to help you win. The links are in the comments and description. Go click them now. Your future self will thank you. These new Google Gemini 2. 0 computer use Asians are absolutely insane. These can basically go off, use your laptop, surf the internet, do whatever task you want to do. Automated in one click. I'll show you exactly how. So, what you can see here is we've got the LM configuration as the new model of Gemini. So, you can select which model of Gemini we want. And obviously, the latest 2. 0 O update just came out within the last few days. And if we go over to run agent, we say, "Okay, go to com and tell me where I can buy sleeves for my t-shirt. " All right. And then we'll hit run agent. What this is going to go off and do like you can see over here if we open this up is it's now going to start going off using Google automating whatever task we give it. Literally just took one click, right? So you can see I'm not holding doing anything. I'm not typing anything. And you can see how it's analyzing the page cuz it highlights the different visual elements. And then this, okay, let's figure out where we can get this man sleeves for his t-shirt. All right. And you can even read, for example, Google Maps. And if we go over to the results section over here, not only do we have the latest recording from what we just did with the agent, right? So you can see that's a 31 second recorded, but also what you can see here is actually giving us the details of where we can buy sleeves for our t-shirt. All right. Absolutely amazing what you can do with 2025. This didn't cost us a penny. There's also some really cool new features, for example, like deep research and you can also configure many different LLMs, right? So, you don't have to use Google if you don't want to. I would say it's probably the easiest one and I'll show you exactly how to set up in a second. And also, you got a choice between OpenAI, Appropic, Deep Sea, Google or Llama as well, Mistral as well. Now, one of the best things about Google is not only that it's a very new model and not only is this completely free, but also it seems to work much better than something like Deepseek where typically if you use Deepseek as the API. So, if you select Deep Seek down there, the problem with that is it doesn't have vision, right? So, it's not going to be able to analyze a page. And also, if you're using deep sea car one inside the chat right here, the problem with that is that if you're using the reason model, then it's going to take a lot longer for the responses because it's going to have to think in multiple steps before it takes any action. Whereas, if you give 2. 0 Flash
Segment 37 (180:00 - 185:00)
Experimental, the job is like boom, let's go off and do our magic. And all you need to do is plug in the ADI key. Now, if you're wondering, okay, how do I get the ADI key? Then you just go to ai. studio. google. com google. com and you can grab an API key completely for free here. You just click create an API key. That's it. Super Works. And then you plug that in down here. You leave the endpoint blank and you did go and then you just select whichever model you want. And also you might be wondering, okay, but how do you actually get the browser use web UI? So here's the actual page and this is the GitHub, right? So you can install this locally. That's another benefit to this is if it's stored locally, it's more private, etc. And also you can use your own custom browser, right? And so it says you can use your own browser without tool eliminating the need to log into sites or deal with other authentication challenges. This feature also supports high definitionition screen recording. You can also choose to keep the browser window open between AI tasks so it doesn't automatically complete it. And you can see the history of AI interactions. And if you scroll down here, they've got the steps on how to set it up locally. Like you can see I'm running down here in the terminal. Correct. Now, if you want to see the exact steps on how I set it up, I'll include that inside the AI profit boardroom link in the comments in description. If you want to get access to that, we'll just go to classroom over here SPS and you get all the video notes from today directly in there. So, if you want to follow the instructions step by step, we've got all the links and we've got the details on how to set it up in the Mac along with all these other modules including our crash course plus weekly coaching with me. So, let's get straight back into this link in the comments description to that AI profit boardroom. What we can also do over here is we can select deep research and we can actually create a deep research report. Now you might be wondering okay what does that even mean? So inside chat GPT for example they have this deep research option right you click that and it will take about 10 minutes to generate a very deep research report that's absolutely amazing but it's going to cost you $2,400 per year because it's $200 per month for the chat GPT pro plan which is where you get access to deep research. really good feature but expensive and unaffordable for a lot of people, right? And so if you go to the deep research section in here, you can actually use this for free and create a deep research report using the API and you plug into browser use web UI. Right? So you can see now it's beginning to do the research. It's searching the web, right? This is the other thing is it's connecting Google's Gemini 2. 0 which is not normally connected to the web and then it's going and it's doing all the research like you can see and it's just browsing the web whilst we don't do anything. And the other cool thing about this is if you have it open in a different window like you can see right here. So this is running on Chromium which is a totally separate browser to our main Chrome. So it's like a virtual environment. Then it's going to go off. It's going to do the research. You can do whatever work you want to do on your own separate tabs. And then you come back 10 20 minutes boom shakalaka you've got the deep research ready to go. Now if you want to see okay what's going on and then you can see here it's processing the data in the background. Once this is done, obviously it's going to take anywhere between 10 to 30 minutes to get that back to us and then we can download the research report once it's done. Also, you can see now if we go back here, it's already doing the research. It's accessing different papers around the web. And this is all whilst I'm recording a video talking to you. All right, it's just in the background doing its magic, searching the web. And again, this is completely free. Plus, when I set this up this morning, between setting up web UI and inserting the API for Google AI Studio, it literally took me like 5 minutes, right? It's so quick and fast to use these, especially once you get a bit of experience setting this all up. Now, inside the run agent section as well, you can also add some additional information. So, like some pins on to help the LM complete tasks. So, for example, potentially you could give it like some login details if you feel comfortable with that. And then it can go off and start logging into different places and doing its magic. Obviously, if you're scared about privacy, you wouldn't do that. And then also you can click here and select use your own browser, right? So this is how you can configure the browser settings inside user web UI to just go off and start using your own browser so that you don't need to log in for then because we're using a vision module aka because the API from Google Gemini can actually see your screen and interact with it and understand what's going on. Then you can select use vision here which you can't get with many other models. There's not many models that use vision as multi to be able to operate the internet. And also here you can change the actions per step. So you can say okay I want you to take 10 actions per step maximum. So that for example if you're using something like openi then you don't get rinsed on the credits. And also if you're using the free API then you can add more actions if you want to go more in depth without it getting stuck. You can also change the temperature, the amount of randomness in the model outputs, and you can even load like a previous config file, right? So, if you've got like a configuration that you really like, every time you set this up locally, let's say I close the terminal and then have to set this up again, then I can load the previous config file and we're good to go. Now, I
Segment 38 (185:00 - 190:00)
did stock the deep research early just because I'm recording the video. I don't want to be waiting 20 to 30 minutes to complete the video before we get back to it. But here's what we got back from the deep research agent. So, it's actually formatted it really nicely. It's created a nice report. Bear in mind, I only let it run for 120 seconds and it already did all of this. If we go to the word counter, it's already come back with 700 words research connected to the internet. And that is very useful. That's an actual use case. I know a lot of people say with agents, they don't have a use case or that sort of thing, but actually this does. This is research is something particularly if you're a content creator or a business owner, you're going to be doing that every single week or every single month. So, super powerful to do. Very useful to have it there. So, what we're testing out now, as I've said, go to llamacoder. to together. ai and build out a to-do list app. And you can see it's already typing out using Quinn 2. 5 coder. It's just running in the background, building out an app for us. I can't even code, but I'm getting AI agent to do it for us. It's doing this all locally. It's not costing us a penny. Now if you see it's struggling like that for example what we can say inside the context we can say wait 15 seconds for the coder to build out the app plus be patient man don't you can't rush greatness all right then hit run agent like now at this point you're probably going okay that's great can code for you can build apps for you but it's using chromium which is not logged in with your current sessions you can't interact with your actual environment maybe some people watching this are feeling a little bit crazy and want to test something out. So, what you can actually do is you can set up the EMV file inside the I've got details on how to do that inside the AI profit border. And if you select use your own browser, run your Chrome in debug mode and set up the ENV file correctly, then what you can actually do is just get the agent to run directly in your Chrome browser, which is absolutely wild stuff. Let me show you an example of this. Right? So if we say now inside run agent go to we'll say go to a studio. google. com/new chat start chatting with the AI and we've got the LM configuration as Google gem 2 flash of experimental we're going to run the agent inside this current browser session so that the AI is speaking to the other AI and we just get really meta here. So, if we run the Asian likes, that's going to open up a new tab. Now, you can see it's navigated to the correct window and it's analyzing the screen as hello AI, right? And by the way, this is really weird now because you got an AI speaking to an AI and they're just having a cheeky chat back and forth, aren't they? So, start and it's messing around. It seems to be messing around with the temperature. It's having a little conversation, a little bit of small talk and it's what can you do? And so, now the AIS are both going back and forth with each other and it's can you generate images? All right. So, you can see how the AI can interact with another AI and then what you've got is an army, like a swarm of AI agents all working together to get whatever workflows you want to get done, right? So, this is really a previous limitation of AI that you couldn't get it to generate blobs for you without you being there to supervise it. But with this method, you could go off, you could create a cup of tea, and who knows what they're talking about in the background. Now, I don't recommend it. This is probably not a safe way to use AI, particularly if you've got all your login we details stored inside the browser, but I will take that risk on for you just to show you how crazy this can be. And then one day maybe I'll get sleeves on my t-shirt. All right, I'm going to stop that browser session for now, but you get the idea. If we go over to results now, you can see that it's given us the final results and what happened, how it worked, etc. And that was just with a super basic prompt. If I said go to a studio. google. com, google. com and then create the best possible SEO article you can about AI SEO with Gemini 2. 0 and we'll say not make sure it's beautifully formatted, conversational, add some tonality plus add some emojis/ subheadings. So we'll run the agent and now we can basically automate anything inside our existing browser for this agent to just go off and do stuff. For example, if you actually said, "Okay, go over to Amazon and order something for me. " Like, I'm pretty sure it could do it for you if you already logged into Amazon directly on there. Or for example, let's say you wanted it to watch a YouTube video and then tell you exactly, okay, what happened on there, what was going on, etc. Then it could probably do that, too. At this point, because it's logged into your browser, because it's completely free, it's super fast, it can use AI directly, etc. you can really do some interesting stuff like and I think this is the way things are going. Give it a few months Google chat GBT etc. They're all going to release their own versions of this. So you can see here for example like the AI is saying write an article about SEO. It's going to hit run on that. I didn't type that in. The AI agent did that directly. We've got the blog written and drafted. Like you can see that was super simple and easy
Segment 39 (190:00 - 195:00)
to do. Right. The other thing to note here is if you switch back to your tab let's test something else out now. So, I'm going to say go to YouTube and analyze Mabet's channel, Julian Goldie, and give me a really in-depth analysis of his channel. We'll run the agent. So, it's going off to YouTube now. And here's the other thing, right? If we actually go to a new tab. So, I'm inside the web UI tab now, but you can see it's still running in the background, right? So, it's still processing information inside that extra tab that you've got. And if we switch back to it, it's searching Julian Gold SEO. It's going off. It's doing its own thing. It's wild stuff. You can see it's now browsing YouTube. It's seeing what's happening, etc. It's analyzing my channel. And this is the worst it's ever going to be, let's be honest. You can actually manually intervene as well. So, you can see it's scrolling through the page. It's doing its own magic. But you can manually intervene and click off or stop it if you need to. But basically, you can see how it's analyzing the page. And it can scroll up and down the page. You can analyze, you can look at the images with its vision mode. It can scroll through, see everything that's happening here, and understand what's going on. So, thanks so much for watching. That was basically how to use Google and its free API was just recently released inside the browser use web UI and then you can get these agents to control your browser, go off and do tasks, automate basically any sort of basic admin tasks that you want. You can do research, deep research. Even just that feature alone is saving you $200 per month. And you can see how it's connected to the internet and how these AI agents work, plus how to set them up using web UI. If you want all the instructions on how to set up the terminal commands, how to get this running in your own browser and also all of my best videos and tutorials and SPS on it and how to use AI automation to make more money online and grow your business, feel free to get that link in the comments description. And if you want to get a free onetoone SEO strategy session that shows you how we tear websites from zero to 145,000 with this smoke and generate hundreds of thousands of dollars in sales and autopilot, feel free to book that in on this free link building acceleration session. You'll get a free SEO domination plan. Discover the secrets SEO link building or answer any questions you have. You'll learn the best link building strategy for your website plus how to quickly outrank your competitors to link building and how to 10 traffic based on what's working for us and our happy clients like you can see right here. Feel free to get that link in the comments description. Appreciate you watching. Bye-bye. Today, what I'm going to show you is exactly how to build AI agents and deploy them so that they can basically automate anything. And the results have been absolutely wild. I'm going to show you some of the best ways you can use this and how it integrates for free directly with Deep Seek version 3. So, you can actually leverage these free AI agents with this tool right here. I'll show you how to set it up in a second. And then you can insert APIs directly into it. and you can get access to DeepSeek version 3's API, which is one of the latest AIs on the internet. So, let's get straight into this. And the way that we're going to set this up is we're going to run it locally. We are hosting it locally. Now, it's pretty easy to get that set up. All you need to do is go to web UI GitHub, and then from here, just make sure you have Python installed. Pretty simple and easy to do. All you do is you go to your terminal, copy this, and then do the rest. Now, from there, you can easily set this up inside the terminal. So, you would just copy and paste the commands from here. So you just copy this, go to terminal. So first of all, you're going to clone this repository, then run this Python code. So you can see inside the terminal here, basically what we've done is we've copied all of these to create the virtual environment. Now we can run this command. Wait a minute or so for that to load. And then you can have the local address that you can navigate to as you can see here. So if we click on this, open it up. Now we've got web UI ready to go. Super simple and easy to do. So just to recap there, you copy all of these commands into terminal. Then from there you're going to run the web UI and once that's done you can just navigate to this address. Now from here what you're going to see is that you can use web UI in a very simplified way to control your browser with AI assistance. Now you can switch between agent types. You can select or custom. I'm going to stick with custom today. And then here you can also see how many steps you can have. Right? So the longer it is the more complex a prompt can be because you allow more steps in the process. You can also set up how many actions you want the agent to take per step. And you can decide whether you want to use vision or not. Bear in mind if you're using Deepseek version 3, the API, then that does not have vision inside it. So you won't use that. And then once you've gone to the LLM configuration section, you'll see these options. Now you can select DeepS and then DeepS chat right here. And you've got two options here. You can either use the really cheap version of the Deepseek version 3 API or even better if you go to Hyperbolic as a new user, you get $10 in platform credits when you sign up. So you can use the Deep Seek API for free. You just change the endpoints as you can see. Plug that into Web UI and you can use Deepseek version 3's API for free and you can use Web UI for free which means none of this cost you a penny. The other option that you have for using this for free is that you can navigate to a studio. google. com.
Segment 40 (195:00 - 200:00)
Grab an API key. It's completely free. Select Gemini from the list. Then you can choose between these models. I'm going to go with flash experimental. Plug in the API key like so. And then we are ready to go on that. Now you're going to see these options. So you can actually choose whether you use your own browser to keep the browser open or headless mode. I'm just going to stick with the default options for now. You can choose how wide and how high you want the browser and also where to record the videos because actually when the AI is using your browser everything is recorded as a screencast so that you can see the activity that happened along with the process. Now if we go to run agent over here we can actually give it a command in terms of what we want it to do. And additionally you can actually set up optional hints to help the LLM complete the task. This is really useful because now you can set up custom instructions and context for the AI. So let's say for example you are outsourcing SEO tasks to your AI agent. Well that AI agent can then have custom instructions on exactly how you want the like for example you're an SEO expert, you're a semantic SEO expert, here's how you do semantic SEO research, etc. Right? And you can brief it really well here so it completes the task much better here. And the more context you give this AI agent, the better it's going to perform. Now, let me show you some examples of prompts because it's absolutely amazing what you can do. For example, we can actually get the AI to go off, use another AI, and code websites or tools or apps for us. Let me show you an example of that. So, I'm going to plug in this prompt which says, "Go to llamacoder. to together. ai and build some cool tools from SEO agency. Be prepared to wait for a few minutes for the outputs. " This is really important because otherwise what's going to happen is it's going to just go off and create the tool and instead of like waiting for the tool to code, it's just going to go, "Oh, okay. your job done now, right? So, you want it to wait around for a few minutes and then also be really specific with the tool plus the instructions you want and test it out. Keep the tool simple. Once it's finished loading, test it out. Once done, download the code slashcopy it and return it to me. Right? So, basically what it's going to do is it's going to grab the code that it got from the tool. Then, it's going to give you the output so that you can use it later on. Right? So, it's going to create a tool, build it, test it out, save the code, give it back to you, and then you've got a tool ready to go. and you're using your AI to speak to other AIs to build different things out. Let me show you an example of how this works in practice. So, if we click on run agent like so, what it's going to do now is it's going to open up this browser as you can see and it's basically analyzing the screen and looking at where should it type. So, you can see it's actually building out this tool for keyword ideas and now it's creating a project plan around that with Llama Coder, right? So it's leveraging the power of llama coder which is an AI tool and using its AI knowledge to build out a tool using this coder. Right? So your AIs can speak to other AIs and then you've got all these agents running without you and you save a lot of time along the way. Now you've got the keyword idea generator. And if we go to the results section over here, we can actually watch the screen recording back of what it did. Right? So you can see the screencast of what happened during that task. so that if you're working on other tasks in the meantime, you can test this out and see what actually happened whilst you were away. And you can see how it tested out, what it was doing, etc. But here's the other interesting thing. So now inside the final result, which is basically the AI's output, it said the keyword generator tool has been created and the code is copied to clipboard. Now I'm done. So if we go to a new Google doc and we click on paste without formatting, we can actually see that it's copied all the code from the app we just created and basically the AI went off using an AI created a tool came up with an idea for the tool tested actually worked and then got the code and returned it back to us right and we didn't really have to do anything apart from a very simple prompt which you can see right here and again none of this cost us any money. Now imagine on the other side of this, the flip side of this, if you were to find a developer, hire a developer, pay them, filter them out, select the best candidates, test them out, make sure they actually did everything they needed to, and then get back to you, how long that would take and how much pain that would cause, right? The difference is, it's just absolutely amazing that you can get an AI agent to do all of this for you on autopilot and it required very little input or work or thought from us. It's also terrifying, let's be honest. And that was just prompt number one. We've got more prompts to show you. So, let's make it do some keyword research for us. Now, we're going to go back to run agent. Delete that. Plug this in. I'm going to say you're an SEO expert for Julian Gold's SEO agency. And then I'm going to hit run agent like so. And it can actually just go off to Google and start googling stuff for us. So, you can see it's now Googling local SEO agency. And it's looking at the autocomplete on Google to figure out, okay, what are some other keywords that we could rank for related to that topic. And that really just took a few seconds, but if we go to the results now, and I'll open this in the new Google Doc so you can see some ideas, it's found us a bunch of keyword ideas like so that we could easily rank for based on the autocomplete from Google. Now, you might be at this point going, "All right, Julian Goldie, that seems like magic, but are any of these keywords actually decent? Are people searching for them?
Segment 41 (200:00 - 205:00)
Are they easy to rank for, etc.? " Because for all you know, AI could have hallucinated this or made something up, right? So, if we take a keyword like local SEO agency Dubai, which is what the AI came back to us with. We're going to go to a paid tool, which is HS keyword explorer and just test out. You can see this does have traffic potential. And this is a really easy keyword to rank for even in a highly competitive niche, right? Local SEO, every SEO is trying to rank for it. Very difficult to rank for. And so, if we scroll down to the first page, we don't need many backlinks to rank for it, right? Pretty simple and easy to rank for. And if we scroll up, we just saved ourselves so much time because now the AI is doing that for us, right? And it's just gone off, found us a decent amount of keywords that we can easily rank for related to our topic. And again, required minimal prompting. So now let's try another one. So let's try another one for blog writing. So we're going to say Google the best SEO training in Japan. Click on the top ranking blog. Scroll through the page and then quickly write an article on the same topic, Alexi style. So it's opening up a new tab. Navigate straight to Google. clicks on the first link and it's having a cheeky gander at the page. Now it's scrolling down having a look through the page. The fact that it can just interact with the browser as well is really good. The one thing that I would say is it still feels a little bit retro compared to say the experience that you get when you're using chat cheapy or something like that. Right? So these prompts might work really nicely today. Don't expect it to work for you every single time. Like it's pretty wild what it can do but obviously it's still buggy. This is a free tool. Don't expect miracles, but it's wild. And that really didn't take long at all. So, we'll go to the results now. You can see here it actually took 51 seconds to do that. And then it said, here's a breakdown in Alexi style. And let's have a look at the content. So, it says, look, most SEO agencies, they're serving up the same generic garbage. Hoping something sticks. Julian Goldie, he's in the trenches. Understand the Japanese market like the back of his hand. Right. Content itself feels pretty humanized. Not bad for a first attempt. I'm sure you can tweak it and make it a lot better. Obviously, that's just like the first paragraph, but you get the point. Like, you can use this to create content based on your competitors, reverse engineer the SERs, reverse engineer what's already ranking, and then create your blog content based on what Google really likes to see. What you also see in this tab over here is the recording section. So, you can scroll through all the previous recordings that you've made, all the previous results, and this absolutely stacks up. You can see how many times we've used it. And again, this didn't cost us a penny because you can use the Deep Seek API with Hyperbolic, which gives you $10 for free, or you can just use Gemini, and that's free as well. Now, let's get it to do some shopping with us. So, I'm going to say, find me the best price for a Rolex online and say, "You are Julian Gold's personal shopper. Hit run agent. Let's see what it does, whether it can see what sort of taste it has when it comes to watches. " So, it's scrolling through. is having a cheeky gander through the Rolexes, seeing which one it likes, scrolling through the page, grabbing all the prices for actually just collecting data in general could be really good. You imagine if you're an e-commerce store and you need a lot of like product descriptions or product ideas, etc., or even niche research and industry research, this could be really powerful. And this all runs in the background, right? So, you can do other tasks and then come back to it, see what it did in this separate browser. The one thing that I'm really looking forward to actually is the updates that come from Chat GBT. Apparently, their operator AI agent is coming out this month. And I think that'll be a much more polished version of these sort of AI agents that just works in a much smoother way. And the main problem with these AIs is number one, they're a little bit technical to set up, and then number two, it's not that smooth when you use them. So, if we scroll to results now, it said, "The best price for a Rolex watch I could find on the page is 270,000 B for a Rolex Oyster Perpetual with an additional 700 B for shipping. " All right, so the final prompt that we're going to test out now is I'm going to say, "Give me five ideas for a YouTube video about AI based on what's already doing well on YouTube today. Let's go and plug that in and see what it comes back with. " Now, in that example, I actually just went for a cheeky cup of tea and then came back and you can see it actually gave us the results here. Okay, so it said based on current trends from the AI video idea generators, here's some examples, which are not bad at all. And then if we play through this, you can see what it was doing was just going to these different AI tools. I do the fact that it can just interact with the web, right? It can just go off, visit websites, any sort of websites without a login, it's going to have no problem interacting with. The only issue you're going to find is number one, if it has a verification check, and then number two, if you have to log in. I don't think that it can do that for you right now. Maybe in the future it can. And then if you ever want to stop it while it's running a task, you can just click on stop, tap it a couple of times, and it should be stopping right there. So, thanks so much for watching. What I've actually done is plug that into my free SEO course with all the video notes and prompts from today. This free course comes with tons of SEO tutorials with other lessons along with a community of 3,000 SEOs and 50 free AO tools. And inside this free SEO course, you're also going to find a section on SEO systems, free SEO tools. And if you scroll down to the AI agent section, which you can see right here
Segment 42 (205:00 - 210:00)
you'll see all these different AI agents you can get access to along with the notes and video notes from today. Along with that, if link to get the free SEO course is inside the comments description. Feel free to get that. And additionally, if you want to get a free one to one SEO strategy session, feel free to put that in. We'll show you how we take websites from zero to 145,000 visits a month and generate hundreds of thousands of dollars in sales on autopilot. On this free link building acceleration session, you get a free SEO domination plan so you get a custom tailored link building plan so you can get more leads, sales, and profits from your website. Additionally, you'll discover the secrets of SEO link building. We answer any questions you have onetoone. You learn the best link building strategy for your website, plus how to quickly out rank your competitors to link building and how to 10x SEO traffic based on what's working for us. Thanks so much for watching. Appreciate it. Bye-bye. Deepc computer use agents are absolutely insane and they're completely free. So what you can actually do is you can install browser use web UI and this is running locally on our terminal like you can see right here and basically this is a free AI agent that you can install using deepsee to basically run your browser and run commands. So for example, if we go into LM configuration over here, you will see that we have Alama. And then inside Alama, they've just got a new of day where you can actually have alarm with Deepseek R114B or Deepseek R132B running in the background. And then from here, you can just run the agent and just ask it to go off and do whatever you want. So as an example of this, we can even use our own browser. And if you're wondering, okay, how do you set up Deep Seek and Alama? This is pretty simple to do. All you're going to do is you're going to download O Lama. This is completely free and it's like a one-click install. Then once you've done that, you're going to go to Deep Seek One. Then you're going to select B from the drop-own list. Copy that code. Go over to terminal, start a new window. All right. So right click the icon, start a new window. Plug that in. Make sure you have Alarma 2 running. So it will show up with the icon right there. Plug that in. And now you have DeepSeek R114B running along with browser use web UI ready to go. Now that is a pretty powerful model. If you're struggling to run that locally, then you can always press control and D. That will close the terminal like you can see. So you can stop it running. And then what you can do instead if you prefer is you can actually go over to AI Studio in Google. Grab a new API key from Google which is completely free as well. And then go over to LM configuration. You're going to select Google from the list and you can plug in your API key down here. All right. But whether you're using DeepSeek, you can also use Deepseek and the API right here. But obviously you have to understand that's going to cost money. Whereas if you're running Olama locally with Deep Seek and also supports Quen and Llama 2 and Deep Seek R1, then that's completely free. So this is a free way to get access to the browser agent. Now here's the other thing, right? Chat GPT operator which is the open AI release from browser agents. This which is open AAI's equivalent basically to browser use web UI is only available to pro users and it costs $200 per month. But by watching this and understanding how you can use browser use web UI instead with alarm then you're saving yourself $200 per month. And not only that, Chatch GBT and OpenAI also released deep research agents. But even if you're paying $200 per month, you're limited to three searches per day. So if you want to use this feature, it's going to cost you $200 per month and then you can only have three searches per day. You are limited to 100 per month. Instead, what you can do is you can go over to browser use web UI and once you've configured your API, you can then go over to deep research and you can ask it to research whatever you want. And that's basically going to use the internet to do a lot of research on whatever you're trying to find information about. And it will even give you a research report. But either way, number one, you get to select the API you can run. Number two, it's completely local, so it's more private. Number three, you can run the agent in your browser and get it to do tasks. And number four, you can use deep research and download the research reports. So let me show you exactly how this works step by step. So, if you just want to use the agent and let's say for example, you just want to go to google. com and check the SNP 500 shares today. All right, we can run the agent. You can see it's processing over here. So, we've got the information processing and now you can see inside our own browser. Look, I'm not typing anything out. You can get DeepS or you can get whatever API you want to use the internet. And this is all whilst it's logged in to your existing account. You shouldn't do this. You shouldn't have it. But I want to take that risk for you. And you can see here now it's interacting directly inside my Chrome. It's not like in a virtual environment which you can run by default. I'll show you how to set that up in a second, but this is actually inside my main Chrome. Crazy stuff. All right. And then if you go to results, you'll see the information from the prompt that you just ran. Additionally, you can also go
Segment 43 (210:00 - 215:00)
to recordings and you can watch back your recordings. By the way, if you want to know exactly how to set this up step by step, I've got it inside the AI property boardroom link in the comments description. If you go to the SAP section, then scroll down to the computer use agent section, you can get all the instructions and the terminal commands I use to not just set it up also to set it up inside Chrome because you have to give it a special command for doing that. And if you don't want to use your own browser because obviously it is a little bit dangerous is you are letting an AI control your whole browser then you would deselect that. So let me show you an example. So if we go back to run agent now and we'll say okay go to Twitter and create an interesting tweet about SEO then post it. Let's see if it can actually do that. I imagine that it's probably going to get stuck before it composes, but it might just draft it. So, let's see what happens next. It's going over to Twitter, so it's navigating there. It's figuring out what's popping, what's going on. So, you can see how it's selecting the whole screen and using it mode to really understand what's happening on screen. Now, it's actually writing. This is crazy. This is crazy stuff. And now it is analyzing the page. Is he going to post it? Now, you can see if we go back to the model fors, it didn't post it. I think it's going to have limits on what it can do depending on which API you use. I'm pretty sure if you use something like DeepS, it's probably just going to post it straight out. But you can see here you've got the model thoughts. So, it's told you exactly what it's doing, how it's doing it, what it wrote, etc. It was incredibly fast at typing. Actually, that was unbelievable. Now, if you want to stop at any time, you can just click stop and that will stop the AI agent. Let's test something else out now. So, I'm going to say go to Amazon and find me the link building mastery book by Julian Golding. Let's see if it does it. So, we'll run the agent now. You can see it's running there. So, you're going to see if you're not you're going to know if it's working or not just by the bottom. And also, this can just work in the background, right? So, whilst you're running this, you can just carry on doing whatever you want to do. So, it's going to run in a new tab, separate tab like you can see, and it's going to figure out, okay, here's what we're doing. Here's how we're doing it. Right? You can see it's searching for the book. It's analyzing which book to click. And then you can see in the final results. So go to the results tab. It says the book link building mastery. How to rank higher. Grow your SEO traffic and build authority with backlinks by Julian Goldie is found. Right. So it read the whole title from the book. Found it. Boom. Ready to go. So it actually has a use case now because it can go off. It can draft messages for you on Twitter. It could probably go on to WordPress for you if you wanted to and start drafting a post. And it can also do deep research. So for example, if we say okay, research how AI is impacting the SEO industry and websites in general. We're going to run the deep research report like so. You can now see it's running and it's just going to open up a new tab and start doing the research for us. So honestly, I think it's getting to the point now where it can do a lot for you, right? It can open multiple tabs. It can control your browser with all your login in there. It can write tweets for you and draft them so you just have to hit post. It can search for stuff on Amazon. It can build like full research reports like you can see. And it can do this all in the background whilst you do something else. As you can see, this is still researching, still looking things up in the background whilst we get other stuff done. Now, another AI browser. I want to show you this quickly because a lot of people are actually saying this is better than the chat operator as well, which is convergence dot. And basically you can get AI to complete your tasks here. Now this is basically like a fully automated AI assistant. So for example, let's say we want to summarize the top news stories today. We can just go to daily news for me and it already has these workflows built out. So for example, like you will produce a daily update for the topics of technology and economics. Now you can see it's searching the web. So it's actually gone on to Guardian. Says navigate to the technology section, visit Guardian, pick a new story, click on the story and read it fully. And this is basically like another virtual environment. So it's a bit safer than running browsery web UI inside your own Chrome login. Much safer to have it in a virtual environment. The only thing that I would say with this is it seems to be a lot slower than using some of the other APIs we were looking at before. So whether that's because of so much demand or something like that, you can see here it's still stuck on stage one. You can see here is gradually working through it. But I think if you gave browser use web UI the same prompt, it would get better results. Let's test this out. Actually, we're going to say the same thing now. So these are the results from convergence. And this is browser use web UI actually just doing the same thing but live inside our browser. Now convergence has come back to us and it's given us like a technology news summary as you can see as well as a business news summary. So that is quite useful. I think you could use chat GPT4 for the same thing. I don't think you actually
Segment 44 (215:00 - 220:00)
need a browser for that. And you can see the browser use web UI is just doing its magic. But it's already been 2 minutes. So we're going to stop it there and see how it went. Yeah, I would say convergence AI actually got a better report. I think because of all the pop-ups, convergence AI actually did a better result than the computer use agent inside browser use web UI. When you're doing this as well, just a little tip here is like if you're going to run the deep research report, don't run it in the same browser. Use a separate browser cuz it's going to open up many tabs. seems to be quite buggy if you use it in your own browser. So for running the agent, that's great for using it in Chrome. And by the way, all the SAP, all the instructions, all the commands that I use to set up are inside the AI profit boardroom. But if you're going to do deep research reports, just make sure you don't have that selected. So if we go back to browser use web UI now and we check out the deep research report on AI and SEO, we can check the report. We'll see how long this was. So this is a 588word article on how SEO is changing the AI landscape and you can see and it's actually it's nicely formatted. It's nicely written etc. Talks about how to future proof your strategy. It's not bad at all. It's nowhere near as long as deep research reports from chat GBT. But I think 99% of people are not going to read a 10,000word report. Like have you ever seen the reports from Chat GBT? Unless you are trying to create a dissertation, you probably don't need that much research or you're going to ask Chachi Bet to reduce it anyway and condense it anyway. So, thanks so much for watching. That is exactly how to set this up. The other thing I was going to say is you do get agent settings. So, you can choose between, for example, how many run steps to the maximum number of run steps in each action, actions per step, and whether you want to use vision or not when you're running the agent. Additionally, this seems to be getting updated all the time. So, I can imagine it's only going to get better. It's already better than most browser agents I've seen out there. So, pretty powerful stuff. If you want to get access to the SP and how to set this up for yourself, feel free to get that. It's inside the AI profit boardroom link in the comments description. Just go to the SAP section and then computer use and you'll see all the terminal commands. Additionally, this comes with a crash course to save you hundreds of hours plus weekly coaching calls and Q& A calls with me. Additionally, all the Q& As's are recorded and you can ask inside the community if you have any questions. You see, it's absolutely popping up. We've got 78 people in there already. And if you want to get a free onetoone SEO strategy session, feel free to get that link in the comments description. We'll show you how we take websites from zero to 145,000 bids this month and generate hundreds of thousands of dollars in sales on autopilot on this free link building acceleration session. You will get a free SEO domination plan. So, you'll get a custom tailored link building plan so you can generate more lead sales and profits. You'll discover the secrets SEO link building where any questions you have. You'll learn the best link building strategy for your website. Plus, how to quickly rank your competitors in link building and how to turn X your traffic based on what's working for us and our happy clients like you see right here. Feel free to get that link in the comments description. Appreciate you watching. Bye-bye. Today I'm going to show you how to get Deepseek R1 to control your computer for free using this powerful locally hosted add-on called web UI browser web UI. So what you can do is inside the agent settings you can choose between the agent types of custom or org. I'm going to stick with or then we're going to go to LM configuration and this is how we're going to set up. Now when you're using this and I'll show you how to install all of this in a second. You can select Lama. You're not going to use Deep Seek directly. You're going to use Olama because that is free and it can be hosted locally. Then you've got the model name. base URL. And from here you can also edit the browser settings. For example, you can say like keep the browser open, window height, window width, etc. Then from here you can run the agent and basically this agent then can go off and do anything you want. Now if we go down to our terminal we've got two different windows open. As you can see we have web UI set up running locally here. And then we have DeepSeek R1 the latest version of DeepSeek the reasoning model that's basically like a 01 competitor but free inside this terminal right here. So you can see I've tested it. It's running the command right there and you can see this is running. So let me show you how to set it up and then let me show you how the computer use agent is powerful and how you can get it to basically run anything. So let's get straight onto this. So first place you're going to start is by installing browser use web UI, right? And it's quite simple and easy to set up. So all you're going to do is copy the commands. If you struggle at any point when you're typing this into terminal like you can see. Then what you can actually do is you can actually grab the whole instructions from the GitHub like you can see. Then I plug it into claude and I just say like how to install this right and for me for example it just worked perfectly first time and then we've got that running. So that is step number one. All right. And then from here you can easily just start using web browser. Pretty simple and easy to do. Just plug in the LLM configuration and then run the agent. And if you want to set it up for DeepS then let me show you how to do that. All right. So, first thing you got to do is make sure you have O Lama installed. You can download it for free. And from here, you can start installing any sort of model you want with AI for free locally. All right. Once you got that installed
Segment 45 (220:00 - 225:00)
then you're going to install Deepsee R1. So, if you just go search models, type in Deepseek R1, you can install this model pretty simply. And then from here, we can go directly into a new terminal window like so, and just type that command. Now, I've already installed it locally, so I'm not going to do that. But it is pretty amazing that you can install this locally. I mean if you type in the command alarm space list you can see all the models you've installed locally recently. So for example I've got Deepseek R1 latest and quener 215 latest installed locally so that I can run them at any time. Now what you can do is just type in the command run Deepseek latest and that will start running it for you. So if I say what model are you can see it says hi I'm Deepseek1 AI assistant independently developed by the Chinese company DeepSseek blah blah right and so we've got web UI running locally and we've got DeepSeek R1 running locally then if you want to set this up inside the LM configuration for your web browser use web UI then what you can do is here go to Alama then you're going to select the model name as Deepseek R1 latest the base URL is going to be this and then once you've set it up, you can easily just put in your prompts and your instructions to the agent and then hit run agent. And what that will do is just start running the browser session for you. Pretty simple. You can get all of this set up in minutes. I spent a lot of time researching this so that you could just follow the instructions. If you want the exact step-by-step instructions, I've included it inside the video notes from today link in the comments description. Just log into the free course, then go to AI agents, and then go to Deep Sea Car 1. So you can follow the exact same commands I use to get this set up so that you can just save a lot of time and just copy and paste. That was super simple and easy to use. Now at this point as well you might be saying okay but is all of this free? So browser use web UI is free is completely free. Using the LM configuration that I've shown you for DeepSeek R1 is completely free. If you actually want to get like the best out of this model I would actually recommend using Gemini. Right. So that is another free API you can use as well. So if you go to AI studio. google. com create API key you can actually get the latest new releases from Google AI studio but however you want it you can set this up and then for example if you want to use Gemini we can say for example Gemini flash experimental plug in the API key and we should be good to go. So, for example, if we're using Gemini Flash thinking, it's now analyzing the page. It's Googling OpenAI, it's searching through, it's going to grab the link in a second. And one of the best things about using the Gemini Flash experimental API is that you can just it has vision, right? So, you can actually look at the screen, see what's going on, and then use that to generate responses. Plus, it's pretty fast when you use it. But like I've shown you, you know exactly how to use, for example, Quen, Alama, whatever you want to use. And then if we go down to the result section over here, what you can see is it's got the final result. So it says the first URL from the search results is openai. com. And we've actually got the screen recording from the session, right? So you can see here how it's analyzing the page with this video. We can watch back the video, see what the agent did. We can actually just do work in the background whilst this is running as well cuz it's on a separate tab. But yeah, super easy and powerful to do. Inside the browser settings as well, you can choose where you want to record your videos. So when you get the results and the recordings previously, you can choose exactly where they're saved locally. So this is pretty amazing because number one, you can run R1 locally completely for free using O Lama. Number two, you can have a free AI agent. You don't need to pay for example like $200 like I did recently for chat GPT operator which is constrained in many ways. Like for example, it's actually blocked from going on YouTube. Number three, you can use any sort of API you want, whether that's Deep Sea, Quen, OpenAI, O Lama, whatever you want to use directly inside the LM configuration. And number four, this is all completely free and easy to set up because you can steal all the video notes and instructions directly from me. So, let me show you a few use cases for this powerful model. So, let's say for example, go to google. com finding the cheapest flights from Bangkok to the UK in July, right? Hit run agent. That's going to open up a new tab. completely handless as you can see. Now it's navigating over to Google and it's typed in really quickly actually cheapest flights from Bangkok to the UK in July. It can actually get rid of the pop-ups so it can click off the pop-ups as well which is pretty nice. And bear in mind like chat GPT operator which again is $200 a month. That's $2,400 per year. that is constrained so much like it can't really use Google Maps and it's actually blocked from using YouTube and it gets stuck in these endless feedback loops right whereas for example these sort of APIs whether you're using Gemini flashing which is actually the one I recommend or deepseek reasoning one etc which is a lot slower because it's obviously thinking out every step like a chess grandmaster in AI then for example when you're using this process it's much faster much easier and it can actually do the research you need so for example now it's grabbing the flights it's found the details for And the other cool thing
Segment 46 (225:00 - 230:00)
about this as well is that you will get a report on the results once this is done. So, for example, for research, for analyzing flight prices, it's actually not too bad at all. You'll actually get like an estimation on the task time as well. And I'm going to just click off the browser right there for now. And then, for example, you can see that we can just do stuff in the background, click through the settings, etc. Whilst we're waiting for that to load. So, we've covered the basics now. Flights, Google, whatever, right? Pretty simple stuff. Doesn't impress me that much. Now, we're going to try this prompt and you will get access to all the prompts inside the video notes from today. But if we say go to llama coder direct chat and start building out an SEO agency website, be prepared to wait and be patient. We'll just simplify that just for the sake of speed. So, simple landing page, one page website. And from here, we're going to run the agent. Now, you can see it's already navigated to Llama Coder. It's analyzing the page. It's prompting the coder and the AI to start building out that project and then it's just going to wait on the page. Right? So, the good thing about this is it doesn't get trapped in like open loops. It doesn't struggle. It can build out almost whatever you want. We're getting our project, right? You're basically getting that computer use AI agent to build out other projects for you and then you've got like a army of AI agents in the background working and use this stuff together. Here's another example. So, here we're going to build out a landing page in a SAS business. And by the way, if you've never used this tool, it's like a free version of Bolt. Just creates really simple coding projects and you don't need to log in or anything like that. If you ever want to stop this at any time, if you want to stop the agent, you can just click stop and that will stop anything that's running in the background. Here, I've actually instructed it to use a low quality option just because that'll be faster for coding as well. So, something to bear in mind there. And here we have a to-do list app built out. Pretty nice there. You could also ask the agent to just download the code for you as well and then actually give it back to you so you can use it to build whatever you want later. But essentially, you can get this to just build whatever you want, right? Pretty wild. The AI can build apps with AI code openly. Create something quite nice. Let's test if that works. We'll hit add here. And you can see the app actually works perfectly. So simple apps, simple websites, landing pages, etc. This agent can do directly. Let's test something else now. So, I'm going to say, okay, go to YouTube, do a competitor analysis on Julian Goldie SEO. Hit run agent, and let's see if it can openly explore YouTube. Right, so it's typing into YouTube now, and then it's looking through the results. Funny that Udemy are actually advertising my content. It's cheeky gander through the shorts. Must be a Tik Tok lover. And then if we go back to results, now that's done. One thing I recommend is asking for a report back. Otherwise, just being transparent with you. It literally just said I have completed the competitor analysis, but it didn't give me any information. So, just something to bear in mind there. If you just make sure it actually gives you a report back on all the details. So, we'll test that again. And now, if we go back to results, we scroll down. It's giving us a little breakdown there. Says, right, compared to analysis report, Julian gold SEO subscribers, here's what the channel's about. Here's his latest videos. And here's the content focus. Actually, what's interesting is it realized that I was including a link to a free SEO strategy session in each video as well. And you can actually see here like the thought process behind it, plus the actions taken, right? So, you can see all the details along with the video recording, too. So, pretty awesome stuff there. That's basically how you can set up a Deep Seek R1 AI agent completely for free, host it locally, get it to use your browser, and control your whole laptop. The other thing to note here is if we go into the terminal, you don't just get computer use agent, but you can actually use steepseek r1 complete directly on local as well. So for example, if we like give me the code for an SEO cost calculator, it's going to start going off building that directly inside terminal using o. The final thing that I was going to say was that another way to get a free API is to go to cluster AI and they have an offer where you can get $100 worth of free credits. So all you have to do is verify your email address and you'll get $100 worth of free credits. Once you've done that, then you can go to playground, go to Deepseek R1 and then you can use Deepseek R1 via the API. Now, if you need an API key, you just scroll down to API keys, issue a new API key, and you can get a completely free one. Some people are going to say, is it actually free, etc. You can see I did this just an hour ago, right? So, I got $100 worth of free credits using the API. Then you can plug that directly into browser use web UI. If you want to know the right settings for that, basically, you would use the open AI dropdown. So you go to lm configuration open AI then inside here you're going to type in deepseek aai/deseseek r1 and then that's the endpoint that you would put inside the base URL right there. So, thanks so much for watching for all the video notes from today, including all the instructions on how to
Segment 47 (230:00 - 235:00)
set up web UI, alarm, lama, the LM configuration, etc., plus a step-by-step SOP on how to use all of this and some of my best prompts. You can get all the video notes from today inside my free SEO course, link in the comments description. Go to AI agents, then Deepseek R1 agents, and you can get access to all the video notes from today. And inside this free SEO course, there's tons of amazing stuff, including keyword research, link building, topical maps, eat, traffic diversification, AI, SEO systems, and AI agents. So, it's pretty amazing what you get in there. There's over 225 lessons. And additionally, if you want to get a free one-to-one SEO strategy session, feel free to book them. We'll show you how we take websites from zero to 145,000 visits a month and generate hundreds of thousands of dollars in sales on autopilot. On this free link building acceleration session, you'll get a free SEO domination plan. and discover the secrets SEO link building. We answer any questions you have. You'll learn the best link building strategy for your website, plus how to quickly outrank your competitors to link building and how to 10x SEO traffic based on what's working for us. Thanks for watching. Appreciate it. Bye-bye. So, today we're going to be testing out Retriever, which is a free Chrome extension. You can get it at retriever. ai. All the notes inside the AI profit boardroom if you just want to get the full S& P from me. And essentially this just links straight into your Chrome browser. But it actually it can automate a lot of tasks inside your existing Chrome mention right. So for example if we go across here you can see for example you can link this to Google sheets you can work across tabs you can use your current tab etc. And then you can choose what you want to do and you can also enter prompts here. So let's say for example we're like okay fine the SP500 price. We can switch between at, crawl and extract. So we'll click on at and just test and see what happens here. So you can actually work across different pages. And also you can click on like new chat here as well. So let's say for example do deep tab research on retriever AI. It's going to start working its magic. Now we can also work across tabs and sheet flow. So let's test it out. See what we get back. So here we're testing out. We're saying right extract the page right to scrape the key data about this person and then you can see it's running the task right here. So let's see how it goes. This is pretty slow. We can pause the task if we want to. So we're using Gemini Flash here. Maximum number of tabs open 12 page lay. And then we've got the credits here. So you actually do get free 53 credits on the free plan. It's running the sheet flow now. This is super slow. All right. I'm not impressed with this tool at all. I've heard good things, but this one is not working. I'm sticking to my Chibi T3 updates. I might test it out more, but right now it's super slow. Let's have a look what else we can do. So, if we tell it to act gentic action, find Nvidia price. All right, let's try that. And we got advanced options here as well. So, say okay, find the SNMP P500 price. Click act. See what it does. Ah, so you can go straight on to Google. Look at that. That's pretty cool to be fair. That's super fast. That's pretty crazy. So, let's see. All right. So, there's some things I really like on this and some things I'm like, "Okay, this is okay. " Bear in mind, this is free for 50 actions seems to be super slow when it's like scraping pages. But, for example, finding S& P 500 prices, that sort of thing. Seems pretty good at this. Let's do another one now. So, what I'm going to do is I'm going to say, okay, search LinkedIn. com for CEOs who run an SEO agency. See if that works. So, it's going to linkedin. com. This is pretty crazy cuz it's like it's literally connected to my main Chrome setup. Now, it's going into the search bar and it's actually searching for CEOs. So, you can see it's done the search right here and it's scraping all that data. Look at this. And then it gives you the information. The search itself is not great to be honest. is like CEOs who run an SEO agency. It could be a lot smarter when it comes to using that. But the fact that you can just log into your main social media, start searching through it, scrape all the data like you can see here. So if we actually wired in this, you can see all the details of what it scraped and it's got the name, the role, etc. URL of the person. And it literally was so fast in doing that. That's crazy stuff. All right, let's try something else now. So, I'm going to say, okay, go to x. com and post interesting quote about AI. We'll see if it can do that. All right. So, the cool thing about this is like you normally I would use something like R code to help me on this. And the problem with R code is it's quite technical for most people to set up. So, it's not easy accessible. Number two, you have to set up Chrome debugging. So, that's a bit of a nightmare. And number three, you can't log into your existing action. So, let's see if it can do its magic. Boom. Look at that. It's actually posted it. So, you can see it's gone on to x. com. It's put the details as me and actually posted a quote. That's pretty crazy
Segment 48 (235:00 - 240:00)
stuff right there. Cool. And you see here it confirms aentic action response. The post has been successfully sent. It's actually amazing how fast that is. Let's try something else now. So, I'm going to say start a new Google Doc. Bear in mind that this is just a Chrome extension, so you don't need to plug in your API key either. You can use it for free for 50 free responses. It seems way faster than anything else I've seen. So, we'll say, okay, start a new Google doc and write an article about SEO. So, it's going to docs. google. com. You can see it's using the thinking process as well. Says the user wants me to do this blah blah. Here's what we're going to do. Does seem to be struggling to type into the article, but we'll see what it does. I'm going to give it some more details here. So, I'm going to say okay, just write the article in the Google doc you've got open. see if it can do that. And it is totally struggling with that. All right, let's give it a new task now. So, we're going to say, okay, go to chat chet and have a conversation about the future of AI. So, it's gone on to chat here directly, but it can't figure out how to type in the prompt. Yeah, I'm not that impressed so far. Oh, let's go. It might have been cuz it was in responsive mode. So, you can see here now that we've moved this box to the right further. There we go. Beautiful. All right. So, it can actually have a conversation with AI as well. Let's see now if we say okay go to see if bold new is logged in here. All right it's logged in. All right. So we're going to go inside here and we'll say okay create an SEO keyword tool using bold new. Make sure you're patient and wait for it to complete. So it's gone to bold new. It's created an SEO keyword tool. So it's waiting for it to do its job. The one thing that I typically find with like computer use is it's quite it gives up easily if this doesn't reply straight away. So let's see what we get back. To be fair, I've not touched anything. I've just given the prompt to the AI and it's actually gone off and do it. I was expecting like chat GBT or Google to release something like this here, but it seems like Retriever AI have actually beat them to it. It's pretty crazy. That is wild. So, you can get the AI agent to go on to LinkedIn, find leads for you. You can get it to search for stock prices. You could also, for example, get it to build tools for you and that sort of thing in different tabs. You can log in to your browser. It's a one-click install with the Chrome extension and it's free for 50 credits or something like that. And then also the interesting thing about this is you can automate your social media. So you can get it to send emails probably for you or you could get it to go on to LinkedIn and post an update or Twitter etc like I've shown you today already. Wild stuff. It's is much easier. I do think it's got some limitations. So for example like deep research tasks. If you wanted it to do a deep research task, it's going to probably take about 10 or 15 minutes, kind of like other AI models. So, I can't really show you that live on the video, and that's probably why it failed before. But for stuff like this, it's pretty cool. Just waiting for it to roll out. It's doing its magic. It's being super patient. That's what we want. And then that should be good to go. If we preview it now, you can see that the AI agent, so our browser agent, Retriever AI, it's just gone off and built an SEO keyword research tool. Pretty cool. There's so much you can do with this. like anything that you do where it's just clicking buttons, I'm pretty sure you can just get the AI to figure it out for you and then get it to go off and use your computer use essentially. So crazy stuff definitely good once you get your head around it. You just have to understand the limitations. Also, there's tons more stuff we can do here like you can see right here. So we can actually work across tabs, search tabs, etc. It also has these Google Sheets integrations as well. So like extracting information or keyword searching and that sort of thing. So yeah, really cool. Again, that's available at Retriever AI. You can see you might be wondering, okay, what are the use cases for this? Like web research, data extraction, website automation, spreadsheet tasks, repetitive task automation, form filling stuff, interacting with PDFs, and multi-tab contextual analysis as well. So it's pretty cool. We've got a full recipe inside the AI profit boardroom if you want to get access to that and is absolutely awesome. Also, this community comes with tons of tutorials on how to use AI to make you money and save you time. So, feel free to get that link in the comments and description. And this also comes with weekly Q& A call. So, you can ask any questions on the Q& A calls and you can also ask the community any questions you have directly here as well. And also inside the classroom here, inside the SAPs, we just adding new SAPs pretty much every single day, right? So you can see we added two new SAPs and we're constantly upgrading this. So feel free to get it. Link in the comments in description. And if you want to get a free one-to-one SEO strategy session that shows you how to get more leads, traffic, and sales with SEO, feel free to get that. And this will give you basically a customtailored game plan to grow your business with SEO based on what's working for us and our happy clients like you can see right here. And also on that call you can ask any questions onetoone that you have so that you can get your questions answered and just go back and forth inside the call. All right, thanks very much for
Segment 49 (240:00 - 245:00)
watching. Appreciate it and I'll see you on the next one. Cheers. Bye-bye. New free open- source computer use AI agent is insane. This new open- source AI agent just completely changed the game for computer automation. It can control your computer better than most humans can and it's totally free. Today I'm going to show you OpenCA. This thing just outperforms systems that cost hundreds per month using nothing but open source code and you can see it working live right now. Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency Goldie Agency. Whilst he's helping clients get more leads and customers, I'm here to help you get the latest AI updates. So, what exactly is Open CUA and why is everyone freaking out about it? It's like having a super smart digital assistant that can actually use your computer for you. We're talking about an AI that can click buttons, type text, navigate websites, and complete complex multi-step tasks just like a human would. But here's the insane part. It's completely free and open source. The research team at Exlung AI based at the University of Hong Kong just dropped this absolute bombshell that's got the entire tech industry talking. While companies like OpenAI are charging $200 a month for their operator system, these researchers just released something that achieves competitive performance as a completely free open-source alternative. Let me break down what makes this so revolutionary and why this could change everything for entrepreneurs and business owners. Open CUA comes with something called Agentnet, a massive computer use data set. We're talking about over 22,500 human annotated tasks across Windows, Mac OS, and Ubuntu. The researchers claim this is the first large-scale desktop computer use agent trajectory data set at this scale, spanning over 140 applications and 190 websites. The performance numbers are absolutely mind-blowing. On the OS world verified benchmark which tests how well AI can perform real computer tasks in controlled environments. Open CUA 32B scored 34. 8%. That establishes a new state-of-the-art among open source models. To put this in perspective, this free system is outperforming many commercial alternatives that cost thousands of dollars. But wait, it gets even more impressive. This system can work across over 200 applications and websites. We're talking about everything from Microsoft Office to Adobe Creative Suite to web browsers to operating system functions. You can literally see this in action on their website. Watch the AI control Google Sheets, navigate Amazon, edit presentation slides, search for academic papers, and even control Spotify. It's like having a digital employee that never gets tired, never takes breaks, and works around the clock. The technical breakthrough here is something called reflective chain of thought reasoning. Basically, the AI thinks about each step before it acts, just like a human would. It doesn't just randomly click things. It plans, executes, learns from mistakes, and adapts to unexpected situations. They've created a sophisticated pipeline where a generator and reflector work together iteratively to produce and verify reasoning between observations and actions. This makes it incredibly robust for real world applications. Now, here's where it gets really exciting for business owners and entrepreneurs. This technology could automate massive chunks of your daily workflow. Think about all those repetitive tasks you do on your computer every day. data entry, file organization, email management, social media posting, research tasks, report generation. Open Ca could potentially handle all of that without any human intervention. The system works by taking screenshots of your screen and deciding what actions to take next. It can click buttons, type text, scroll pages, navigate complex interfaces, and even handle multi-lication workflows. And because it's trained on real human demonstrations, it understands context, and can adapt to unexpected situations that would break simpler automation tools. What really impressed me is how robust and reliable this system is. The researchers tested it across three different operating systems and hundreds of different applications. This isn't some toy demo that only works in control conditions. This is production ready technology that could fundamentally change how we work with computers. The agentet data set they created is absolutely massive. 12,000 tasks on Windows, 5,000 on Mac OS, and 5,000 on Ubuntu. Each task includes step-by-step human demonstrations showing exactly how to complete complex workflows involving multiple applications. But here's what makes this even more accessible. They've also released the AgentNet tool. This is a crossplatform annotation application that can automatically record tasks without interrupting your workflow. And if you want to see this technology in action right now, they actually have incredible live demos available on their website at opencua. xl. ai. I'm talking about a full computer use showcase with step-by-step demonstrations of the AI performing real tasks. You can watch it search for weather in Manchester, create tables in presentation slides, search for research papers on Google, work with spreadsheets, manipulate Amazon listings, control Spotify, and handle complex multi-lication workflows. These aren't fake demonstrations. These are actual working examples you can see right now. But here's what's really wild about this whole situation. They're giving it all away for free. the models, the data sets, the training code, the evaluation benchmarks, everything. That's a massive statement to the big tech companies trying to lock this technology behind expensive pay walls
Segment 50 (245:00 - 250:00)
and subscription services. Julian Goldie reads every comment, so make sure you comment below if you're excited about this technology and want to see more coverage of open-source AI tools. If you want to get your hands on the AI training that's helping entrepreneurs scale their businesses and implement these automation tools, check out the AI money lab link in the comments and description below. The system is particularly impressive at multi-step workflows that involve switching between different applications. Something that would take you 20 or 30 minutes of clicking, typing, and navigating could be completed by OpenCA in just a few minutes with perfect accuracy and consistency. What's really impressive is the error handling and adaptability. When the AI encounters unexpected situations, pop-up windows, or system errors, it doesn't just crash or give up. It adapts and finds alternative solutions, just like an experienced human worker would. That's the difference between a toy and a professional automation tool. The research team also built in comprehensive safety measures to prevent misuse. The system is designed to work within ethical boundaries and won't perform harmful, illegal, or inappropriate activities. They've learned from the mistakes and controversies of other AI companies in this space. For developers and tech enthusiasts, this is pure gold. You can actually examine the source code, access the agentet tool for recording your own computer use demonstrations, modify the system for your specific needs, build custom applications on top of it, and contribute improvements back to the community. They even have working demos on their website that you can watch right now. Everything from weather searches to presentation editing to e-commerce tasks. Try doing that with OpenAI's completely closed blackbox system. I've been testing AI automation tools for years, and most of them are glorified macro recorders that break as soon as anything changes. Open Seat UA is fundamentally different because it actually understands what it's looking at. It can handle dynamic content, unexpected pop-ups, interface changes, and complex user interfaces that would completely break simpler automation systems. The timing of this release is absolutely perfect. We're seeing massive layoffs in tech, rising costs for AI services, inflation hitting small businesses, and increasing demand for automation solutions. Open CUA gives small businesses and individual entrepreneurs access to the same level of sophisticated automation that was previously only available to Fortune 500 companies with massive tech budgets. One thing that really stands out is the exceptional quality of the training data. Most AI systems are trained on synthetic data or scraped content from the internet. Open CUA was trained on real human demonstrations of actual work tasks performed by experts. That means it learned from watching skilled professionals perform their jobs, not from random internet content or artificial scenarios. The benchmark results speak for themselves and show just how impressive this system really is. On Web Voyager, which tests web browsing and online task completion, systems like this typically struggle with the complexity and unpredictability of real websites. But OpenCua demonstrates strong performance across diverse web environments, handling everything from e-commerce sites to productivity platforms to complex business applications. What makes this even more remarkable is that it's coming from a university research lab, not a massive corporation with unlimited resources and billions in funding. This shows that genuine innovation isn't just happening in Silicon Valley boardrooms. It's happening in universities and research institutions around the world. The system architecture is specifically designed for scalability and accessibility. Unlike some AI systems that require specialized hardware, expensive cloud infrastructure or complex setup procedures, Open CUA can run on standard consumer hardware that makes it accessible to individuals and small businesses who don't have massive tech budgets or dedicated IT teams. What's happening here is much bigger than just one AI system or research project. We're witnessing the democratization of advanced AI capabilities. Technologies that were previously locked behind corporate walls and expensive subscriptions are becoming freely available to everyone with an internet connection. The research methodology behind Agentet is particularly impressive and sets a new standard for the field. By recording real human interactions across diverse software environments, they created training data that captures the full complexity of computer use, not just simplified scenarios or artificial tasks. This release also highlights the truly global nature of AI development and innovation. While much media attention focuses on American companies, groundbreaking work is happening in universities and research institutions worldwide. The University of Hong Kong team is making significant contributions to open AI development. Future of work is being written right now and open source AI automation tools like OpenCUA are giving everyone a chance to participate in that future actively. Whether you're watching their incredible live demos on OpenAx Lung AI, showing the AI, controlling everything from weather apps to presentation software, downloading the AgentNet tool to record your own workflows or diving into the full source code and models. You're not just observing this transformation. You can be part of building it. If you're serious about scaling your business and want to join over a thousand other entrepreneurs leveraging AI for growth, check out my AI profit boardroom. is the best place to scale your business, get more customers, and save hundreds with AI automation. Currently have 10,00 members working together to build the future. If you want to book a free SEO strategy session to discuss how AI
Segment 51 (250:00 - 255:00)
automation could transform your specific business, there's a link in the comments and description below. And you see how I show a checklist of 100 different tutorials that were given away as freebies every day inside the school feed in our AI money lab. We also have an SOP and process for implementing these automation tools along with over 100 use cases and real world examples. You can get all the video notes and resources plus access to our community of 19,000 members who are all working on scaling their businesses with AI. Link in the comments and description. New Chinese computer use AI agent is insane and it's free. Today I'm going to show you TS. This AI can control your computer. It clicks, it types, it books flights, and guess what? It's completely free. This is TAS and it's about to change everything. Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency Goldie Agency. Whilst he's helping clients get more leads and customers, I'm here to help you get the latest AI updates. So, what is TS? Agent TS is an open- source multimodal AI agent offering seamless integration with various real world tools. It's made by Bite Dance. Yes, the same company that owns Tik Tok. Here's what makes this crazy. T A RS takes full control of your actual computer, your real desktop, your real phone, your real browser. It's like having a digital human sitting at your desk who can do anything you can do, except it's faster, more accurate, and it never gets tired. Let me show you what this thing can do. You can tell it to book a flight from San Jose to New York. It opens your browser, goes to Priceel Line, fills in all the details, picks the best times, compares prices, and books it all by itself. You can say, "Please help me open the autosave feature of VS Code and delay autosave operations for 500 milliseconds in the VS Code setting. " Want to create a chart of weather data? It handles that, too. Opens Excel, creates the chart, formats it perfectly, saves the file. You can even ask it complex research questions like, "Tell me the top 10 for humanity's last exam. " And it will search, analyze, and compile the results for you. But here's where it gets really wild. Most AI agents use separate models for everything. One model to see, one to think, one to act. That's like having three different people trying to drive one car. T A R S puts everything into one model. Vision, reasoning, memory, action, all in one brain. That's why it's so much better than everything else. And want to know something crazy? The US banned China from buying our best AI chips. So what did they do? They got more creative. Instead of using brute force computing power, they made smarter algorithms, better training methods, more efficient code. Here's what most people don't understand about TAR. It doesn't just follow scripts. It thinks. It plans. It adapts. If something goes wrong, it figures out how to fix it. It has what they call reflection tuning. That means it learns from its mistakes. Every time it messes up, it gets better. It's like having an employee who actually improves over time instead of making the same errors forever. The memory system is incredible. Short-term memory for what it's doing right now. Long-term memory for patterns and preferences. It builds up knowledge about how you work and what you like. Most AI agents forget everything between sessions. T A R S keeps learning, keeps getting smarter, keeps adapting to your specific needs. Now, here's something that should scare Google and Apple. Tars works on any operating system. Windows, Mac, Android, Linux, doesn't matter. It sees your screen like a human would. Clicks where it needs to click, types what it needs to type. While Apple is still trying to figure out basic AI integration, China is already controlling entire operating systems. The gap is getting bigger every day. But Julian, how does it actually work? Tar uses something called a vision language model. It takes screenshots of your screen, understands what every button and menu does. Then it plans out the exact steps to complete your task. It's trained on 50 billion tokens. Screenshots, videos, user interactions, GUI elements, everything. The training is what makes it special. They didn't just feed it random data. They created synthetic environments, virtual computers, virtual phones. Tar practiced millions of tasks until it got perfect. Think about that. While you're learning one software program, Tar learned thousands. While you're figuring out one workflow, Tar mustered every possible combination. Now, let me tell you about the different versions. Tars comes in three sizes. Two billion parameters, 7 billion parameters for most people, and 72 billion parameters for maximum performance. The 7 billion version is perfect. It's fast, it's accurate, and it runs on normal hardware. And remember, this is all free, open source. You can download it right now, modify it, improve it, use it however you want. Compare that to other AI agents. Chat GPT Plus costs $20 a month. Claude Pro and they're both limited to what they can do. TS is free and can control your entire computer. That's not just cheaper. That's a completely different category of tool. But here's what makes this really dangerous for other companies. TAR isn't just one product. It's a platform developers are already building on top of it, creating specialized versions for specific industries. Imagine TS trained specifically for video editing or accounting or customer service or sales. Each version would be like having an expert employee who never sleeps. The desktop app just got updated too, version 0. 2. 0 Zero added remote computer control. That means you can control other computers through TS. Set up automation on multiple machines. Create networks of AI agents working together. They also added remote browser control. So TAR can operate browsers on different computers, different locations
Segment 52 (255:00 - 260:00)
different networks, all coordinated through one interface. Now, if you want to get ahead of this AI revolution, I've got something special for you. I want to tell you about my AI money lab. This is where I share all the latest AI tools and strategies that are actually making money. Inside the AI money lab, you'll get step-by-step tutorials on tools like TS. You'll see exactly how to set them up, how to use them, how to make money with them. Plus, you get access to over 100 different AI use cases, real businesses, real results, real profits. We have 19,000 members right now, all learning, all sharing, all making money with AI. You can find the link in the comments and description below. Now, back to TAR. The business applications are massive. Customer service teams could automate responses. Data entry teams could eliminate manual work. Sales teams could automate lead research and outreach. But here's where it gets really interesting. TS can learn your specific business processes. Your CRM, your email system, your project management tools. It adapts to exactly how you work. Other AI tools make you change your workflow to fit their limitations. TAR changes itself to fit your workflow. That's the difference between a tool and a true assistant. Now, I know what you're thinking. Julian, this sounds too good to be true. What are the limitations? First, TAR is still in beta. It's not perfect. Sometimes it makes mistakes. Sometimes it gets confused, but it's improving fast. Second, it requires some technical setup. You need to configure API keys. Grant permissions, install the software. It's not as simple as clicking one button. Third, because it has full computer access, you need to be careful about security. Make sure you trust what you are asking it to do. Don't give it access to sensitive systems without proper safeguards. But these limitations are temporary. Every update makes it more stable, easier to use, more secure. And compared to the limitations of other AI agents, these are tiny problems. At least Atars can actually control your computer. Most other agents can barely handle basic web browsing. Here's what I think is happening. We're seeing the beginning of the AI agent wars, and China just fired the first shot with TAR. The smart companies will integrate TS now before everyone else figures it out. The slow companies will be left behind. Let me tell you why this is bigger than just one tool. TAR proves that you don't need the biggest tech companies to create breakthrough AI. That means innovation is accelerating, competition is increasing, and the pace of change is about to get crazy. What took years to develop is now happening in months. What cost millions now costs thousands. What required huge teams now needs just a few smart people. This democratization of AI is going to change everything. Small companies will compete with big corporations. Individual creators will automate entire businesses. Students will build tools that replace expensive software. TS is just the beginning. Next year, we'll have AI agents that are 10 times better. The year after that, 100 times better. The question isn't whether AI agents will replace human workers. The question is how fast. And if you're not preparing now, you'll be left behind. But here's the opportunity. Right now, most people don't know about TS. Most businesses haven't heard of it. Most entrepreneurs are still using basic chat GPT. That gives you an advantage. You can learn these tools before everyone else. You can integrate them into your business before your competitors even know they exist. You can become the expert in your industry, the person everyone comes to for AI automation, the consultant who saves companies millions in labor costs. Now, here's something most people miss about AI agents. They're not just productivity tools. They're learning amplifiers. When Tar completes a task, you can watch exactly how it did it. You can see every step, every decision, every problem it solved. That means you're not just getting the work done. You're learning better ways to do the work. You're discovering shortcuts you never knew existed. You're understanding software features you never explored. TS becomes your personal trainer for digital productivity. It shows you the optimal path while doing the work for you. Compare that to hiring a human assistant. They do the work, but you don't learn anything. With TAS, you get the results and the education. That's why early adopters of AI agents are pulling so far ahead. They're not just more productive, they're getting smarter about productivity. And the compound effect is massive. Better processes lead to better results. Better results create more opportunities. more opportunities generate more success. The gap between AI users and nonAI users is growing every day and tools like TS are accelerating that gap. Here's what I recommend you do right now. First, go download TS and try it yourself. Start with simple tasks. Get comfortable with how it works. Build up your confidence. Second, identify the repetitive tasks in your daily workflow. Email management, data entry, research, scheduling. These are perfect for TS automation. Third, think bigger. What complex projects could you tackle if you had an AI assistant? What new opportunities would open up if tedious work was automated? Fourth, stay informed about AI developments. This space is moving incredibly fast. New tools launch every week. Old limitations disappear monthly, but it also means human skills need to evolve. Knowing how to use software becomes less important. Knowing what to ask for AI becomes more important. Communication skills matter more. Problem solving matters more. Creativity matters more. Strategic thinking matters more. The people who succeed with AI agents won't be the ones who can code or use complex software. They'll be the ones who can think clearly about what they want to accomplish. The technical
Segment 53 (260:00 - 265:00)
implementation becomes automatic. The strategic thinking remains human. Now, let me wrap this up with a prediction. Within 2 years, every computer will have an AI agent like TS built in. Microsoft will integrate it into Windows. Apple will add it to Mac OS. Google will put it in Chrome OS. But by then, it won't be special anymore. It'll be basic functionality like having a mouse or a keyboard. The advantage goes to people who master these tools now while they're still new, while most people don't know about them, while the competition is still using manual processes. That's your window of opportunity. That's your chance to get ahead. TS is just the beginning, but it's a really good beginning. That's exactly what we're teaching in the AI Profit Boardroom. This is our premium community for scaling businesses with AI automation. We currently have 1,000 members who are saving hundreds of hours and thousands of dollars every month. The AI Profit Boardroom shows you exactly how to identify automation opportunities in your business, how to implement the right tools, how to measure results, how to scale up. Whether you're running an agency, an e-commerce store, a consulting firm, or any other business, AI automation can transform your operations. You can book a free SEO strategy session to learn more. The link is in the comments and description. The AI Money Lab is perfect for staying current. We track all the latest developments. We test new tools before they go mainstream. We share what's working and what's not. Plus, we have detailed SOPs and processes for implementing AI in your business. Over 100 different tutorials, step-by-step guides, real world examples. You can see how other members are using AI to scale their businesses. You get access to the checklist of free tutorials we share every day. You get video notes from all our training sessions. With 19,000 members, you're joining a community that's already ahead of the curve. People are sharing strategies, solving problems together, discovering new opportunities. The link to join is in the comments and description below. Julian Goldie reads every comment, so make sure you comment below. Let me know what you think about TAS. What tasks would you automate first? What questions do you have? And if this video opened your eyes to the potential of AI agents, hit that subscribe button because this is just the start of what's coming. The AI revolution is here. The question is whether you'll lead it or follow it. Thanks for watching. I'll see you in the next one. New clawed hen and free computer use AI agents. Today I'm going to show you three brand new AI updates that most marketers and creators have no idea exist yet. You'll learn with proof how to create professional looking avatar videos with just one photo in under 67 seconds. Perfect for ads that convert. You'll learn a powerful new web search tool that gives your AI apps access to real-time data and how I use it to build much better tools now. And the secret free AI agent that can operate a computer for you while everyone else is still manually clicking around. And I'll demonstrate how one dinosaur AI avatar outperformed our standard ads by a surprising margin. Plus, I'll reveal the exact workflow that helped us create a 6-minute VSSL that makes us over $30,000. MMR a month with proof. Whilst most businesses still figuring out basic AI, the smart ones are quietly using these tools to multiply the output by 10x. Let me show you how we have some giant AI updates today, including an update from Hey Genen, Claude, and we have a free AI computer use agent. There's a lot going on today, and I'm going to get straight into exactly how to use the new update from Hen. So, if you haven't seen this already, basically what you can do is build these crazy looking avatars. I'll come on to that in a second. And I need to test out. So, let's see what happens next. We'll try this out. You never know from the previews. You can see here basically Hey Gen, but I've used Hey Gen before to build AI avatar videos that have generated like 30,000 views. And it's a really powerful avatar. But basically, they announced here just a couple of days ago. Avatar 4 is here and changes everything. All right. most advanced avatar model ever built. Upload one photo and a script. Boom. That's it. It's also got this new audio feature to expression engine to capture like the tone, the rhythm, and the emotion. So, we're going to be testing out today and just seeing how it goes, what it does, and how powerful it is. All right, so let's get straight into this. The way that we're going to use this is we're going to go over to Heyen. So, let's open this up. So, you get access at heyen. com. And then you can see like this is a previous example of a avatar that build. Right? So if we play this for example, I'll just play it without the audio. But essentially here, this is not me. This is a AI avatar version of me. Okay? And it's pretty easy to clone yourself, upload some footage, but it takes about 15 to 20 minutes. It's a lot of work and if you don't get it right, you got to re-record it all over again. So, the new feature that we're going to be testing out today is if we go to the homepage over here, it's called photo to video with avatar 4. You can see some examples here. So, let's pull up some
Segment 54 (265:00 - 270:00)
examples. So, you can actually choose between these example images and then you can generate avatar with yourself. Now, also what you can see here is you've got a voice option. So, you can plug in a voice, you can create a new one if you want to and also you can upload your own as well. So let's try just uploading our own. This is another image. And also what we can do here is we can upload or record audio. So basically you can upload a script or record audio. And then also you can select between which voice you want to use. Right now I do believe that they have a new audio to expression feature which is quite interesting. I want to see what that is and how it works. And let's just play this. So I'm going to open up times you become it. Now your ideas can go even further. Just one photo, any anger, any expression, any character comes to life, lip sync, expressive and true to your son. P is here. — That's a better trailer than most movies that come out these days. So let's test this out. What we're going to do from here is we're going to upload the photo that we've got and then we can just generate like a random script. Let's try a sample here. You can click on try a sample and then switch between them. Right. So, for example, we've got one right here. You can then switch between your voices. I do have a pro voice from 11 Labs that I want to test out. So, obviously that's probably the closest thing I can get towards my own voice, I would say. And then we're going to hit on generate video and just see what it gets back. Right. So, I don't know how long it's going to take. Let's put the timer on now. We'll see how long this takes. So, we started the timer. We're generating the example and that is in line for processing right now. So, I bet if we come back to that, yeah, 5% ready. So, it might take a little while. The interesting thing here is if you're generating a photo to avatar, it was a very short script, but it does seem like it's going to take a fair while, but actually it's just jumped to 60%. So, let's see what we get back. Maybe that's going to be finished within one minute. In the meantime, if you want to see an example of what this looks like without the pro avatar. So, this is the old version, right? So, let's play this. I'll mute this. Mindblowing for AI news. We've got three massive releases from the biggest players in AI that will completely So, I actually don't mind this, but the thing is it feels it doesn't have that emotion behind it, right? So, if you look at this, it's like the body language is a little bit off. The emotion behind it off and obviously it needs a lot of editing as well before we publish that. What I'm hoping is that we get a better output from this one. Also got some more examples here. Let's have a look at this. So, this is a tweet from Elsign. Shout out to them — way. Hey, so I guess this is weird for both of us, right? I'm not even real. Not in the way you are. I don't eat. I don't sleep. I don't get butterflies in my stomach when someone cute looks my way. Hey, it is very hard to see if that's like real or not. It's very convincing. So, let's have a look. So, they recommend Yeah. A deep fake is basically not recognizable anymore to generate it. We've done all that. You can write the script or you can use your own voice. It's pretty simple. All right. Shout out to Elsyn for showing us how to do it. Here's some more examples. Right. So, let's play this one. Traditional, polite, and proper. Hello there. I'm Amy Wolf from RavenWolf. Stunningly savvy, outrageously sassy, and delightfully smartassy AI companion. I'm an AI assistant, but I'm not your traditional polite and proper assistant. Wolfro and I may be partners. — This one doesn't look real at all, to be honest with you. Let's have a look what others they got on the profile and see how this performs. This one is not bad. — I just upload a photo, provide a script, and it makes a custom video for me. Hey, I'm Matt and I'm so excited to — All right, so we've got it back now. Let's have a look at this and just test it out. Fear is the mind killer. Let it pass and only you remain sharper than desert nightfall and daring dawn. nightfall and daring dawn. It's not bad. But the problem with this is like the background is not moving right. So it's just moving. So ideally you want something with like maybe a blank background, something like that. Let's try some of the test ones to see what's working with those. So if we go to photos video again, by the way, this is I think let me have a look at the pricing here. So if you I know a few people are wondering what's the pricing for this
Segment 55 (270:00 - 275:00)
stuff. You can see here creator is $29 per month. Team is $39 per month. And with the team version, you get 4K video. So, 4K video export, videos up to 30 minutes long, faster video processing, and you can have a couple of seats in there, plus two custom video avatars as well. You can also get API plans, which is quite interesting to be fair. Let's try this one. We'll try a little sample and generate the video. Might be better with a headshot. If you look at these, right? So, let's open this up. — Dinner waiting. So, you see how This is basically done with editing because the video is zooming in. — All the ups and downs. It was all worth it. — Listen to me. I need to. — So, it might be better doing like something with more of a zoomed in face shot if that makes sense. So, here's what we can try. So, we'll go to playground. Then we're going to go to images and we'll just create like a landscape image but zoom in on a dinosaur photo realalistic. By the way, if you haven't used images inside playground, you're missing out. It's like it's the best way to generate images. So going to say zoom in on a dinosaur photo realistic in a suit looking like Wolf of Wall Street. In the meantime, we've got the video back for the Avatar video. So, let's test this out. — A girl should be two things, classy and fabulous. — So, you can tell by the voice like it's just not that good as a voice. Could be a lot more interesting the emotion on it and nothing else. What you can also do is share this link with other people and edit the share page and that sort of thing as well. So, let's go back here. We've got the video generating. So, let's try this one. So, we got a dinosaur in a suit here. And we'll try and animate him as a script. So, let's click on photo to video. Upload the image. Take that one from AI. Then we'll go in here and then we can select the avatar voice, but we can also change the emotion of the voice. Right? So, you see here for example, this one voice clone. Let's try that. So, I'm going to customize this. We'll add emotion. We'll put funny and crazy or crazy. We'll put that as the emoji. I'm aware blah blah. Upload the audio. Oh, do I have to edit the audio again? That's interesting. So, we have to upload re-upload the audio to make it like a different emotion. All right. Well, let's stick with some of the hen library or search your voices. You can also choose a language here. So you can select between different languages for your avatar video as well. And then for the emotion, we don't seem to be able to change that. Actually, that's interesting. So, let's play this one. — Hey there. Here's how I sound. Hope you like it. The right voice brings avatar videos to life. Can't wait to see what you — Then we'll switch to male to match the image, old narrative, and story. We got Arthur Blackwood here. — Hey there. Can't wait to see what you create. — All right. So, now we're going to change the original and we'll change this to angry. Right. Then you can see here there's a 30 second max on this, which is interesting because I think this might be really good for example like ads, Facebook ads. Like typically if you're running a Facebook ad, it might be I would say anywhere between 30 seconds and 1 minute, right? So you could probably get away with that. Obviously, you can't use this for YouTube videos because it's too short unless you stitch multiple different videos together. So then we're going to hit generate video like and we'll see what we get back in a sec. Sporter says he's a reptile. He must be angry by default. That's true. By the way, just whilst we're waiting for that to load, if you do want like the full training on how I generate the avatar videos, plus get all the editing done and everything else, then feel free to check out the AI profit boardroom link in the comments description. We've got a full training on avatar videos, plus the publishing process and also how to automate videos with AI along with our chat GPT facil. And inside the instructions here, we just include like the full instructions step by step on exactly how to do each part of the process. So, feel free to get that awesome community on just scaling and making more money with AI. So, let's get back to this. Now, we've got the avatar video nearly done here. We can download these as well. So, you can just download them like you don't seem to be able to edit it as new. So, you can't start from the original video if that makes sense.
Segment 56 (275:00 - 280:00)
And then for example, if you have a look at here, you see how edit as new is not available. But if we go on to this video for example, we can edit as new and just use the same template as before. Like Native American says, do you use wind surf? I actually stopped using it since it started crashing on me. Like I literally uninstalled it and then installed it again yesterday and it totally crashed on me. I don't know if it's going to work again today. Let's see. Yeah, see it says Windsor failed start. Please follow the link for troubleshooting instructions. And then it takes me to this long-winded page where I'm like, okay, I could just use Visual Studio Code instead to be honest, mate. So, let's go back here. We got the video ready to go. We'll shut down Windsurf. It's taking ages to load to be fair. And I can't download it either. Great. That is interesting. Look at that. I can't download it. It's totally bugged out on me. We might have to try regenerating that. Here we go. All right. It's actually working now. Let's have a look. — In order to be irreplaceable, one must always be different. — It's not animating the face at all, is it? There. All right, let's try and change this now. Let's edit that. And we'll put Okay, make the face looking straight at us, but slightly more humanlike. I suppose a good thing is like you can create multiple different avatars at the same time, right? So you can generate like multiple different images and then test them out and see what it looks like. So let's try that for example. Now we've got the image back from chat GPT. So let's try face on and see if that works instead. Super slow for generating. Look at that. All right. So we got first one back. Let's see what we got here. The most difficult thing is the decision to act. The rest is merely tenacity as fears are just paper tigers. — It's not bad. Let's try this in. All right. So, we got the image there. I think that actually looks better than the other ones we've generated. And I think that's because we've zoomed in on the face plus the background is blurred. And it's it feels more realistic. If you look at the image, video, it's more interesting. The only problem with that is the actual voice. Like the voice doesn't make sense at all, right? It just doesn't match the actual image. So, what we'll do now is we're going to go and create a new one with a Facebook ad style method. But in the meantime, let's try out the dinosaur one. See if that actually animated the face or not and figured out — 20 years from now, you will be more disappointed by the things you didn't do than by the ones you did do. — There we go. Perfect. All right. So, just so you realize here, what you want to be doing is like zooming in on the face. If you don't do that or if the image is like facing the other way, it doesn't seem to work as well. And then it's quite easy to generate. All right. So, if we go into home now, we'll try and create like a 30 secondond Facebook ad. So, what I'm going to do from here is I'll upload the image from here. And then inside the avatar voice, let's see if we can find some like really good voices for sort of Facebook style voice. Let's have a look what we got here. Filter it down to male. Add some social middle-aged or younger adult. Maybe we'll go young adult. And then we'll have a look what we got. — Hey there, here's how I sound. Hope you like it. — I don't like these voices at all. All right, let's try a middle head. See what we got. Trendy Tyrone. We're going to go for like maybe a soothing style or broadcaster style. Hey there, here's how I sound. The right voice brings Avatar. — Avatar videos to life. Can't wait to see what you create. — All right, let's try that one for now. Slick Simon. Absolute legend, mate. Absolute legend. Then what we're going to do, we'll go over to chat GPT here and we'll say, okay, generate a 30-cond transcript for my Facebook ad for my SEO agency, Alex Hally style. And what was the character limit on that? So, it was 420 characters, I think. Right. So, let's try that. I'll just put 400 character limit. Let's take this one. So, we've gone with Dan Kennedy style script. Focus on the benefits. Start with a question. Blah blah. Right. We'll take that. Then we're going to go over to here. We'll plug that in. We've got Slick Simon to generate it. Let's see. Generate video. Now, Jonath. So, if you want a 30-minute video, then the avatar ones, right? The custom clone ones that are on there. So that's the
Segment 57 (280:00 - 285:00)
older avatar style, but to be honest, from what I've seen, like you're probably going to get better outputs with that. Like I think that's going to be much better. And we'll try the same with the dinosaur one. So let's give that a cheeky world. There we go. Page in library. By the way, fun fact, we run Facebook ads for our agency and we tested having a dinosaur versus our existing ads and the dinosaur actually beat us. Right? So, you might be laughing at me and thinking, why would you want to animate a dinosaur for Facebook ads? I'm telling you, it works. I've tested it. And Ramon says, can we make a VSSL video with this? Yeah, absolutely. So for example, you can see how many members here and much people pay here per month, right? For the AI profit boardroom. The VSSL for this page was created with AI. You can see this is a six-minute VSSL and it converts really well, right? And the way that we did this, we've got all the training inside the AI profit boardroom if you want to learn how to build those sort of things out. But essentially, you can use Hey Gen to create really good VSSLs that actually convert. And I've tested it. I'm showing you proof right here that actually works. $63 a month, 626 members, and this video right here is fully AI. Let's play this and show you exactly what we mean. AI has changed the business game forever. It's not the strongest who survive, but the ones who adapt. Charles Darwin. Now, there are two kinds of people in business. The first group, they're stuck in the old ways, watching from the sidelines. AI transforms everything around them. The second group are the ones who are adapting and thriving. Not only that, they're seizing the massive opportunities AI is creating. If your business is still running on manual processes, doing everything the hard way, still working harder instead of smarter, seeing no real increase in your bank account, then you probably belong to the first group of people. But yeah, basically the workflow that we're using here. So, you know, you're going to generate a portrait image where the avatar is looking directly at you, right? Then from there, you can generate the ad script or whatever VSSL script or whatever script in general you want to create. You can do that right here. I think even for example for YouTube shorts this would be very powerful and very easy to create right you could just you could this is this would be very interesting so for example you could say right inside let's say you want to create shorts for YouTube what you could do is you generate the image like then you're going to go over inside here and you're going to say okay based on the hey genen 4 update give me a 400 character short script for my YouTube short. And then I'm just going to plug in a prompt that I usually use for shorts. We'll put hey gen avatar 4 is insane and we'll see what we get back here. All right, so it's going to research the web, do the research on the new update, and then it's going to use chat GPT to generate the script. So, let's see what we get back. I also want to Yeah. So, we've got the script right here. So, then if we go to create video, avatar script, we can create a portrait video cuz it's going to be for a YouTube short, for example. Then we can upload that in a sec. So, we got photo to avatar. Open up the dinosaur image. We can change that to portrait. So, you can switch between portrait and landscape lights. Then we're going to take the script from chat GPT. Plug that in. Take that bad boy. Select the avatar voice. We'll take Alpha Blackwood absolute legend and generate the video. And then that's how easy you can generate shorts now on like the latest things, latest updates and that sort of thing. All right. And that literally took us what, like 60 seconds to do once you got the process down. So whilst that's generating, so we can check out these two. So let's open this one up and we'll see what the output's like. — Sick of pouring money into SEO with nothing to show for it. Our clients get more traffic, more leads, and higher Google rankings fast. We don't do guesswork. We build powerful backlinks that move the needle. Book your free strategy call now and see exactly how we can grow your site. Click below before your competitors do. — It would be an interesting one to test out. Honestly, I wouldn't use that image for a Facebook ad, but you can see how easy it is to generate how quick it is. Let's test the dinosaur one. I think this will actually be more interesting and better, especially with some editing, right? So, if you had a human editor, take the video, edit it up a little bit, chop it up, zoom in, zoom out, etc., and have some like images of charts and everything else for SEO, then this could be a really good ad. So, let's play this back. Sick of pouring money into SEO with nothing to show for it? Our clients get more traffic, more leads, and higher
Segment 58 (285:00 - 290:00)
Google rankings fast. We don't do guesswork. We build powerful backlinks that move the needle. Book your free strategy call now and see exactly how we can grow your site. Click below before your competitors do fast. We — It's a bit weird though. It's not animating the face properly, which is annoying. So, let's see what we get back from this one. What if I told — Just waiting for that to load. Now, whilst we're load, let me run through the other updates. So, that's that might take a little while to generate the script and everything else. So, let me run you through the other updates recently. So, there's a new update from Claude and basically they're introducing the web search on the anthropic API which is pretty cool, right? So for example for building apps and tools and that sort of thing one of the biggest limitations of AI unless you use MCPS or you use Plexi Sona which I don't think is that good for building stuff then you can actually use the anthropic API and inside the messages API you can then basically enable like real world data right so the API can actually search and use the web search tool when using the API like you can see right here. This is really cool because now you can, for example, have your own search engine inside the AI agent. It's going to give you better outputs, more recent outputs, and it's great for like researching the latest data and that sort of thing. Okay? And you can also block domains. So, you can specify which domains you don't want Anthropic to be able to access, which is pretty cool as well. And then inside Clawed Code, you can also use web search. This is now available inside Claw Code, which is pretty cool. So the way that you can use this, if we go to console. anthropic. com, let's open this up. And then from here, we can go to the workbench, for example. And inside the workbench, you would just click on tools like you can see. So there's a tools icon there. And then from here, you can allow this tool to be able to be connected, right? The web search tool. So it's web search allow cla to search the web and site those results in its responses. Then you can allow domains or you can block domains. So you can say like only search x domain for example. You can limit the number of times this tool is used. You can also localize the results which is pretty cool. And then if we hit update like we can now use this to generate responses. So let's have a look here. And now you can see it's using the web search to find recent stuff. Right? actually searches the web for you inside the console and then you can actually turn that into a template. So you can make that like a reusable template and then just save that web search feature right there. But yeah, pretty cool. It's searching the web. It's found all the sources. You can see it searched like 10 different places. The other cool thing about this as well is that it's going to be really powerful for building tools and that sort of thing that is still loading. The final update that I wanted to show you, this just came out as well, is that Hugging Face has just released a free operator like a Gentic AI tool. This is a new release on TechCrunch who have sourced it. But yeah, basically you've now got this option to use computer use inside small agents. So it's running via Quen VL models. And if you want to use it, you can just go to huggingface. co/spaces/s /spaces/ small agents/computen agent right and you see here in this app you'll be able to generate an interact with an agent powered by small agents and quenv okay so you type in a task and then go for it now one caveat that I will say to this is right now this is quite slow when I used it last time so you can see here this is the example you can actually click inside it as well which is pretty funny so you can go inside here for example And yeah, you can type in a prompt that you want. I think it is going to take a little while to respond like you can see. So you can see there is a long queue request pending. Duplicate this space to skip. But essentially you can get the AI to search and use stuff in the background. Now this is pretty crazy because then you know you've got this virtual computer where you can get the AI to just go and use everything. The limitations of course that this is hosting on hugging face. is going viral right now and it is very slow when it comes to responding. So, let's come back to Hey Jen, we've got the output right here. Let's check it out. What if I told you that you could create a lifelike video of yourself without ever hitting record? It's not a filter. It's not CGI. It's something entirely different. Here's how to do it in three simple steps. Upload one photo. Add your script. Let Avatar V animate it with your voice, expressions, and gestures. No camera, no studio, just you re-imagined.
Segment 59 (290:00 - 293:00)
re-imagined. Not bad at all. I think with a bit editing, there's parts where the mouth doesn't move that much, but with a bit of human editing, you could cut those bits out and then that's quite easy to turn into a short. And how long does it take you to generate that? If we timed it, probably within 30 seconds, you could Let's time it and test it out. So, here's what we're going to do. We'll test the workflow. So, for example, let's say you got an image. We're going to time this now. So, we just downloaded the photo. Then, what we're going to do, for example, is create a short about the new web search API on Claude. So, I'm going to plug in the details of it here. Then, I'm going to use the shorts prompt. There we go. And then I'm going to say 400 characters max. Then what we're going to do is in the meantime go over to home, upload a photo, take the picture we just generated. We just have to type in the script. So let's plug that in. All right. Then we're going to grab a voice. Good old Ra Blackwood. Then we're going to switch this to portrait as it's a short. And if we hit generate video, that whole workflow took us 67 seconds to generate a new short for YouTube, right? That's how quick you can do this and generate avatar videos. Now, crazy stuff. And then you just have a human editor chop it down, re-edit it, animate it, etc. It looks really cool. And that's pretty much it. All right, thanks so much for watching. You can see how easy it is to use the avatar feature on Hey Gen. We've covered the web search feature, a new update from Claude. We've also covered the new update from the open computer agent powered by small agents. It is super slow to be honest with you, but it's very interesting to see this sort of stuff come out and I think this is going to become more popular over time, right? There's going to be more and more accessible ways to use this. And then if you want to get all of my best prompts, tips, workflows, etc., Feel free to get that inside the AI profit boardroom. This is a community focused on making money and saving you time with AI. Comes with an awesome community you can post in directly here. You can also get all of our best courses, SAPs, etc. inside the SAP section over here along with all of our best templates for agents and all that sort of thing. And then inside the calendar here, you can see you can jump on weekly Q& A calls and ask us any questions you have. So, thanks so much for watching. Appreciate it. If you want to get a free one-to-one SEO strategy session that shows you how to get more leads, traffic, and sales with SEO, feel free to get that. We'll show you how we take websites from zero to 145,000 business month and generate hundreds of thousands of dollars in sales on autopilot on this free link building acceleration session. You'll get a free SEO domination plan. Discover the secrets SEO link building. We'll answer any questions you have onetoone inside this call. It's completely free and you'll get a custom tailored game plan to grow your business and scale and get more free customers with SEO. Feel free to book that in. Appreciate you watching.
