# HUGE AI NEWS: OpenAI New BREAKTHROUGH? Stunning New AI Image Model, New Robotics Company And More

## Метаданные

- **Канал:** TheAIGRID
- **YouTube:** https://www.youtube.com/watch?v=hACwBNRR3to
- **Дата:** 04.12.2024
- **Длительность:** 27:09
- **Просмотры:** 31,551
- **Источник:** https://ekstraktznaniy.ru/video/13632

## Описание

Try Wondershare Virbo which will help creators and business owner scale content creation and save costs: https://bit.ly/4ihz0Iu 

Prepare for AGI with me - https://www.skool.com/postagiprepardness 
🐤 Follow Me on Twitter https://twitter.com/TheAiGrid
🌐 Checkout My website - https://theaigrid.com/


00:00 OPENAI 
02:18 AI AGENT DEVELOPMENT
04:05 AMAZON AI SECURITY
05:26 AMAZON'S AI MODELS
07:43 AMAZON CHIP DEVELOPMENT
09:18 CHINA AI COMPETITION
10:44 LUMA LABS PHOTON
13:09 ELON MUSK DEVELOPMENTS
16:12 TESLA & ROBOTICS
18:44 MAGIC LAB DEMOS
20:06 GOOGLE & 11 LABS UPDATES
23:43 CHATGPT NAME MYSTERY
26:04 OPENAI ANNOUNCEMENTS

Links From Todays Video:
https://www.reddit.com/r/singularity/comments/1h47d8k/in_the_last_67_months_amazon_saw_an_increase_from/ 
Sam Altman AI Will be co-worker - https://x.com/tsarnick/status/1863367799772754000/video/1
Luma AI
https://x.com/LumaLabsAI/status/1863688635780042795 
Google Veo 
https://x.com/googlecloud/status/1863957264895217999 
https://cloud.googl

## Транскрипт

### OPENAI []

so one of the first things that actually happened was something at openai now this video is rather interesting because Rune a for you know well not former a potential open ey employee there's various so one of the first things that actually happened was something at open ey now the details are vague as per usual if you know anything about open ey drama but we did get this cryptic tweet from run uh potential openai employee there's not much information confirmed and also Mark Chen someone that is an SVP research at opening ey so this is a really interesting tweet and I actually said to myself like what do you mean that open ey is back I didn't even know that they were gone you know so I decided to tweet this I said what happened I didn't even think open ey were gone and run responded saying to be honest I think we had just an extremely fun party now I think that is a funny comment maybe they did have a really fun party but it's quite possible that they did have some other information going on so we actually got this information that OPI managed to poach three top Engineers from Deep Mind so if you're not familiar with Google deep mine that's the company SL you know business behind Google's Gemini models and essentially the reason that this is pretty incredible is because we do know that when we take a look at all of the AI research that exists in the space a large majority of the Innovative research has come from Google AI engineers and research researchers which is incredible because of course now that we have three you know top leading computer vision researchers you know Lucas Bayer Alexander zaha all three of these individuals have left Google Deep Mind to join opening ey and this is you know incredible because they were behind state-of-the-art Vision approaches and open source models like vit sigp and P Gemma so this is a huge boost for opening eye in terms of talent because they're able to get really smart individuals on their team and that does mean that now open ey at basically housing some of the world's best talent now this is something that is rather fascinating a later in the video I'll talk a little bit more about openi and some of the things that they have said and recently we did get some mman that was actually talking about how a agents are going to work in the future as its

### AI AGENT DEVELOPMENT [2:18]

capabilities go up um maybe right now you can give it a 5-second task without supervision um and eventually you give it a 5 minute task and then a 5 hour 5-day task and maybe someday can go do a 5mon task and that's like a you know full scientist off exp point something but I think it'll feel more like that an increasingly senior coworker not one moment it was not autonomous and then and I think Sam Alman stating that look you're going to be able to do a 5day task or five month task I think that is something that is pretty ridiculous not ridiculous in the sense that it's not going to be possible but ridiculous in the sense that this is crazy that we're starting to talk about this as a possible reality in AI before I can continue talking about the crazy stories in AI today let me tell you about a tool that's changing the game for Content creators today's episode is sponsored by wondershare verbo as someone who spends hours creating AI news content I know how challenging video production can be wondershare verbo is an AI Avatar video generator that's caught my attention because it lets creators make videos without ever stepping in front of a camera what's really interesting is how it can transform any text into video content using over 300 lifelike AI apps avatars you can even create talking photos that bring still images to life perfect for those who want to keep their content faceless while maintaining a professional look for my International viewers here's something cool verbo can translate your videos into different languages with perfect lip sync offering over 460 voices across 90 countries they've even integrated chat GPT for scripts generation which could save hours of writing time like if you want to try it yourself verbo offers 2 minutes of free video credit for new users whether you're thinking about starting a Channel or looking to your existing content check it out at verb. wondershare. com now let's get back to today's mind-blowing developments now we

### AMAZON AI SECURITY [4:05]

also did see that Amazon unfortunately is seeing hackers you know one of the things is that AI makes a lot of things really you know easy and it lowers the barrier to entry you can see right here the Wall Street Journal interviewed someone at Amazon and they said how many attacks are you seeing these days and CJ Moses says we're seeing billions of attempts coming our way on average we're seeing 7 50 million attempts per day previously we'd see about 100 million hits per day and that number has grown to 750 million over six or seven months so this is something that is absolutely insane and you know Wall Street Journal asks is that a sign that hackers are using Ai and Moses responds without a doubt generative AI has provided access to those who previously didn't have software development Engineers to do these things and now it's much more ubiquitous such that normal humans can do things that they could before and they can just ask the computer to do that for them so that is something that I think is going to be an unfortunate reality for the future we're largely going to get a lot of these hacks that are going to be there now when it comes to Amazon that isn't the only thing that Amazon are in news for they're actually in the news for a lot of good things which is pretty good considering Amazon haven't had that much to do with the AI space in terms of releasing their own foundational models but of course now we

### AMAZON'S AI MODELS [5:26]

get to see exactly what Amazon have done so recently Amazon I released a video earli today on my channel talking about how Amazon have literally joined the generative AI space in terms of their Foundation models and they're actually firmly wedged within the AI ecosystem they released three models that were you know Nova Pro you've got Nova light and they've got Nova micro essentially a pro model that is quite like you know Gemini Pro in terms of you know where they're trying to place the reasoning capabilities quite like a GPT 40 scale model of course you've got Nova light which is essentially a much smaller model and a much more lightweight model and then they of course have Nova micro which is of course a tinier model for other sort of use cases that you're able to do which essentially just means use this model for a lot of quick use cases where you just need basically an LM to verify certain things so you can see here that the Nova models are the ones in the orange so you can see Nova Pro here you've got Nova lights here and you've got Nova micro over here so overall you can see that this model is right next to claw 3. 5 it's right just above mystal large 2 and llama 3. 14 5B and of course you can see for the flash models the Llama 317b GPT 40 mini that's where Nova light is we can see for Nova micro it's quite like these 8 billion parameter models now like I said before I released a video on this earlier today you can go and check that out but it is really interesting to see Amazon getting into the AI stage at this stage now I think it's important for Amazon to do this because you know they have a lot of things that they're probably going to use these models for of course they've got Alexa other Amazon products that they probably might need to use these models for so developing Your Own Foundation models is really good because although they do have a really solid partnership with anthropic it makes sense for them to develop their own foundational models so they can actually control certain things that they want to so that is something that I think is rather important for them now another thing that I made a video on as well is that Amazon are going to be developing their own chips which is their trinium two chips so that they can train these things as much as possible and so that they don't have to rely on Nvidia and they actually talk about how they're going to be basically developing this super cluster which is going to be used to train Claude and many other AI models

### AMAZON CHIP DEVELOPMENT [7:43]

this is something that we're building together with our partners anthropic uh and it's actually going to be several hundred thousand chips um uh and they're focused on and it's our new tranum 2 chips and so we expect that this should deliver them five times more compute uh than they've used in their last Model training and so we're really excited about what they're able to accomplish and um and they're expecting to build much bigger uh and much more capable AI models from uh that large compute cluster we built our own chip actually and so we've it's called tranium 2 um and we're quite excited about the performance that we get out of that and that's combined all in these really large Ultra uh clusters that combine uh 64 trainum chips all to deliver 83 pedop flops from a single node and so that is the first Innovation is our Amazon designed custom silic that gives us um really unparalleled performance for generative AI capabilities and then we build these together with high performance networking that we also build in house so I think yeah when we actually take a look at what Amazon are doing in the space here I think it's going to be really interesting to see how they manage to speed up the kind of ecosystem because not only do they have their own models but clusters where they're able to train models quicker faster and more efficiently so it's going to be something really interesting to see now so we actually had an interesting interview where these individuals were regarding the China situation and that situation is where a lot of these companies SL countries are really competitive in nature and right now there is this big elephant in the room with as to whoever gets the AGI first is basically going to control the entirety of the future of the human race because if you get to AGI you get to ASI

### CHINA AI COMPETITION [9:18]

and basically once you get to the ASI you basically have powers over anyone else that's not me saying something crazy that was actually said by a former opening ey employee and basically in this interview they were discussing you know what that means and how China is accelerating due to open source models being freely available a question you should ask is what is China's greatest AI accelerant today and the answer not from us from people inside the AI field say it's meta it's America's companies releasing open- Source AI is causing China to go faster yeah are you kidding no and just in a senate hearing um uh recently uh open AI insiders revealed that uh basically they were not appropriately locking down and securing their AI models why would they not do that so it's like China actually sees open AI as maybe one of the biggest accelerance of their progress if you can steal the model so it's like the Manhattan Project like we're developing this very powerful technology it's only as safe as it is secure if you just leak the secrets that's we're not going to beat China by running faster and leaving the door wide open for someone to steal it Luma Labs AI introducing you know surprisingly introducing their new photon text image models now this is absolutely incredible because essentially what they've done here is they've managed to introduce a model that is text to image and the reason that is crazy is because they are originally a video company so I wouldn't have expected them to come into the text

### LUMA LABS PHOTON [10:44]

image game and even if they did I wouldn't have expected them to make a model that is not only better than anything that exists on the market but also a model that is cheaper than anything that exists on the market so this is crazy because I remember people were stating that like every you know months or so every piece of AI technology or tool is literally going to be close to zero in terms of cost and literally every month we see that there are new tools new things released that are not only better but also cheaper and faster so it's one of the craziest things I've ever seen like if we're just taking a look at the aesthetic the instruction following the composition the overall we can see that Luma Photon ranks levels above mid Journey above flux 1. 1 Pro above you know recra V3 above IDR V2 and above you know stable the fusion 3. 5 large so this is something that I think is remarkably incredible considering the fact that when we take a look at how fast things are moving we have companies that are literally out competing each other every month I remember when flux 1. 1 Pro was released everyone was like there's nobody that's going to defeat them and then of course we had other companies then we had I ideogram V2 and now we've got Luma Photon so this is going to be something that's absolutely insane and we actually also decide to look at the cost we can see here that this is 1. 6 cents for a 1080p image and 0. 4 cents for a really fast image so it's going to be absolutely insane in terms of what we're able to do with these models at scale in terms of content Creation in terms of just a wide range of creative uses so um we also do get this video here which they show you exactly how you know it is able to adhere to your prompts so I think that this is something that is really effective because prompt adherence is something that's not discussed enough and I remember when I first got into the AI image generation space I was like okay a lot these images look cool but they don't really state slash do exactly what I want them to do now we actually get some Elon Musk news here and this is actually some great news for Elon Musk so Elon Musk actually got a really nice deal with Nvidia I'm not sure how he managed to land this but priority for NVIDIA gb200 delivery in January with $1 billion so Elon musk's company has managed to secure chips for his you know company and this is actually crazy

### ELON MUSK DEVELOPMENTS [13:09]

because I remember that it was Sam mman that was actually in an article you know not in an article but it was in like a Microsoft meeting and allegedly he was stating that you know Elon Musk is literally getting too much compute like how is he able to get so much compute this quick Microsoft what is going on and apparently he was you know not Furious but just quite agitated that he could get so much compute in such a short time considering the fact Microsoft have billions of dollars they have all those relationships but Elon Musk somehow managed to you know secure these chips secure the compute before you know samman had so this is something that's you know pretty incredible here of course we also have the fact that Elon Musk also decided to file an injunction at open AI essentially trying to prevent them from doing a wide range of things and you know I guess it's understandable I covered this in another video but the gist of this is that Elon Musk doesn't want open ey to continue performing the way it is because number one safety researchers have left number two you've got the situation where open ey and Microsoft are you know doing these anti-competitive practices number three openi is trying to get this you know Monopoly where they are basically saying to people look don't invest in xai when you're investing in open AI you're only allowed to invest in us and of course number four the fact that opening ey completely abandoned its original Mission of being a nonprofit company and is now trying to switch its for profit status so it seems that Elon Musk is really trying to stick it to samman I also got to see this tweet where you know someone breaks it down it says that you know Elon Musk has managed to bulldo sound's political attempts has first access to the new Nvidia gpus and already has the biggest cluster and is tying up opening eyes move to for profit in court and owns all the real world coner streams for example you know he owns Tesla which has humanoid robots he has autonomous cars also owns X this social platform and someone says how does Sean win here and you can see you know he says openi has likely already started a Manhattan project with the US government and is training on government user data which is probably quite accurate considering the fact that you know there are some people that you know are deep within the you know USA government that actually work at openai but of course you have to remember Elon Musk was recently I think he's recently hired as the department of government efficiency for the United States so it's going to be really interesting to see how that works because Elon has firmly wished himself as an ally of trump and of course now that Trump's the president that is a lot of power so it's really interesting to see the power plays that are going on and of course if you aren't familiar with what's going on between Elon Musk and Sam mman there has been a long going feud between those two ever since Sam Alman has decided to make that company for profit now in more Elon Musk news Elon Musk basically just confirmed that with grock 3 they are basically going to address they're basically going to add all court cases to the training set so Gro will be able to render extremely compelling legal verdicts this is going to be super interesting because we know that there are some companies out there that are working on this kind of thing but of course it seems that grock 3 apparently is going to be the best model ever it's not me that's

### TESLA & ROBOTICS [16:12]

saying that but Elon Musk earlier this year did state that grock is going to be one of the best models in terms of pretty much everything and it should be released later this year I'm not sure I'm going to hold them to that timeline because as we do know Elon Musk is rather optimistic with his timelines in terms of anything he releases and of course with AI not just with Elon musk's company but we've seen numerous slowdowns in terms of the timelines for certain models for example claw 3. 5 Opus was supposed to be released Gemini 2 was already and unfortunately we haven't managed to see anything just yet so it seems like we will see what this month entails when it comes to AI releases now one of the craziest things that I personally think was just absolutely insane was the fact that we had the Tesla bot that was able to to perform this catch not once but twice now I know that most people like I saw that there was a comment on my video that had 100 likes and it was like until this is you know autonomous I don't really care and I don't think that that's the attitude that people should have I know that yes this might be teleoperated but I think it goes to show what is possible with the current robotics Hardware which means that like if this was completely autonomous do you know how advanced these robots would be and I think this you know needs to you know really let you understand where these robots going to be in the future because if this you know right now is tell operated in 3 years something like this might not be teleoperated and if we have robots that are able to catch things out of midair you know this is like a lot better than a lot of people would do like you think the average person can catch a ball out of the midair quickly and you have to think about it like this if we're able to have these robots Mass scale perform one software update now they can jump now they can back flip now they can punch fight like Bruce Lee do you know 10 different billion things I mean you have to understand what the kind of impact that has on the world so I think just saying you know ah you know it's teleoperator y y I completely get the skeptical fact that you know companies try to wow us with these amazing Tech demos and then they never deliver on the final product completely understandable but regardless or not whether it's this company or another company it seems like this robotic future is going to be there and of course one company that is doing that is Magic lab and they recently showcase this demo where they actually showcased a fleet of humanoid robotics that are existing and then they just get up and go to work so I think this demo was rather surprising because we got to see how a fleet works together as a team and how they manage to perform certain tasks

### MAGIC LAB DEMOS [18:44]

in harmony so we got to see these robots be able to pick up certain things able to perform certain tasks we've got to see them interact with the real world and the craziest thing about this was that this company they're not working in stealth but they don't really perform regular updates in terms of what they showcased us like they'll just come out with a new video every 6 to 7 months or so and they'll just showcase what the robot is able to do so this is something that I think was super interesting because we got to see the robots actually walk a decent speed I mean if I actually rewind it quickly I'm going to show you guys exactly what I mean right here so you can see that like on the left hand side we can see the robot is just walking pretty normally and then on the right hand side we can see it's walking quite fast because it doesn't have anything in its hand so these robots are a little bit more advanced than some that currently exist but it does seem like they have some sort of simplistic Hardware that focuses on the raw details needed to achieve the tasks so I'm really excited to see what comes out of China because I wouldn't be surprised if you know like they literally just come out with a bunch of humanoids that fast anything that the US has to offer simply because right now that is their main focus and when it comes to factories and efficiency they are really hard to beat now we also had 11 Labs actually released something that is a competitor to Google notebook LM if you aren't familiar with that tool that's essentially a toour that allows you to convert books and PDFs into podcasts and as for Cinderella

### GOOGLE & 11 LABS UPDATES [20:06]

and the prince they lived happily ever after there is always more to every story and this is a brand new way to experience it introducing our new Gen FM podcast feature say hello to your hosts I mean seriously I find Cinderella just so inspiring as a character throughout this whole book yeah I know right and she does it all in those tiny glass slippers unreal now you can generate your own podcast all at the click of a button all right so let's dive into today's review topic atomic theory now this is one of my absolute favorite subjects where should we begin choose any topic any voices anytime next the recipe says to lay out the apple slices until the pen is full ah yes this is a true classic of fine French cuisine personalized podcasts for everyone around the world and all in 32 different languages Gen FM now available in your pocket with the 11 reader app now we actually also got something pretty cool from Google I know that most people don't really pay attention that much when Google releases stuff but they actually did release something that was really cool and that is of course their vo model so I'm not sure if you guys know what that is but vo is image to video generation and they release this on their Google Cloud platform so if you're someone that uses this is basically something that you can now utilize so you can use whatever image you want and it's really high quality image to video now I got to be honest Google should have got this out a lot earlier because so many other companies have surpassed them in terms of what they able to do but I still think that Google has a decent reputation and of course they have the largest distribution that exists out there I mean you've got people that use Google search you YouTube it's quite likely that they're going to embed this into those platforms so maybe they don't really care that much but it is pretty interesting to see the kind of tools that Google are working on that still hold up to what the AI standard is in terms of you know the things that we're looking at from other companies like you know openi s Runway and of course Luma laab so this is going to be something that I think of course if you want you can experiment with it but I don't think Google are going to release this as a standalone product of course it's in their AI Labs which is basically where you can test a range of different Google AI products but I think this is going to be something that is probably just you know blended into something like YouTube shorts or potentially like a YouTube Studio editor so when we take a look at the examples here like you know a slow motion of marshmallows roasting over crackling fire looks pretty cool you know people dancing at a festival also looks pretty cool we can also see the prompt of an initial closeup of like a bear playing a guitar and of course a ship floating on the sea so overall this does look really interesting and I wouldn't be surprised if we saw this sometime soon so then of course we had this really interesting piece from someone on Twitter that discovered that chat gbt refuses to say the name David mayor and nobody knows why so basically this was where if you actually try and ask Chad to be to say the name David Mayer in any sort of conversation it would just print an error and it would just say I am unable to produce a response which is kind of interesting because there aren't many things that would get chat GPT to just completely like you know just produce an error like that of course you know

### CHATGPT NAME MYSTERY [23:43]

asking it to produce something illegal is probably going to say you know I'm not able to talk about that but this is something where it literally can't say one person's name that of course there was all sorts of Investigations with as to who David mayor is of course people are really trying to you know figure out why that isn't the case and you know who this person is which is I guess you could say kind of ironic considering the fact that this is probably someone that doesn't want that kind of attention being put on them so I guess that is kind of ironic in a sense but what we do have here is we do have an actual look at what openi have said so it says open AI had confirmed on Tuesday that the name David Meyer has been flagged by internal privacy tools and saying in a statement that there may be instant where chat gbt does not provide certain information about people to protect their privacy and the company would not provide further detail on the tools or process so basically this is something that confirms that there are certain individuals that have looked at you know chat GPT and they said look okay I don't want my name in this yada y and I'm guessing that maybe there's some kind of private service that openingi has where they're like look if you want your you know name to not be mentioned in the software this is something that we can do or maybe there's just certain things that they don't want I know that there are other names as well I'm not sure what their scenario is but it seems like this is something that is official coming from open AI so that solves that mystery now of course we also had Dario emid actually talking about his P doom and he actually said that he doesn't even like the term P Doom anymore what's your P Doom these days I don't really like you've learned your lesson on this I never like those I never liked those words I think they're kind of weird I I just like my my view is just that like look we should measure for risks as they come up uh and you know in the meantime we should get all the economic benefits that we could get and we should find ways to measure the risks that are effective but are minimally disruptive to the amazing economic process that we see going on now that you know the last thing we want to do is slow down now one of the next things that we had here was we had opening eye actually talk about you know 12 Days 12 live streams a bunch of new things big and small the 12 days of open I starts tomorrow so this twet is currently growing viral I just saw this after I made this video and I wanted to

### OPENAI ANNOUNCEMENTS [26:04]

include it here because it's really interesting because we have a situation that basically says that look over the next swap days open AI are going to show us exactly what they've been working on in a variety of different demos so this is pretty crazy because it's one of the things that I've wanted to see from opening eye for quite some time and we actually take a look at what's going on here it seems that this recent tweet by the open ey employee if you remember at the start of the video how I spoke about how you know run and Mark Chen were basically stating that look opening eye is back and you know you don't know why this actually makes a lot of sense when we take a look at what opening eye are saying and 12 days for 12 demos is a ton of different things so I'm not sure what it could be maybe it's going to be a sore release you know the full advanced voice mode but this has the AI industry holding its breath right now because the last time opening a I did a demo it really did shock the world so I'll be waiting I'll be there I'll be making a video if this does happen and if you enjoyed the video I guess I'll see you in the next one