NEW Google Agentic Vision Update!

Today we're going to be looking at the new update from Google. So Google have released aic vision which is a new frontier AI capability in Gemini 3 flash that converts image understanding from a static act into an agentic process. What that means in English is like if you have a picture of a workflow that you want to implement, you can now give it to aentic vision inside Google and then Google can turn it into a workflow, an agentic workflow which is pretty insane, right? So basically how this works is we have think, act and observe, right? This is the loop that works with a gentic vision. So step number one think the model analyzes an image query then architects a multi-step plan. So that is step number one. Top of that the model will then generate and execute Python code to actively manipulate or analyze images and then step number three is observed. So the transform image is appended to the model's context window allowing it to inspect the new data before generating a final response to the initial image query. Let's test out and see what that actually means in English. So you can see here for example it says Introducing Agentic Vision inside Gemini Flash. Agentic vision, a new capability combines visual reasoning with code execution to ground answers and visual evidence, right? And Gemini 3 flash is a really powerful model. I use it a lot inside anti-gravity, which I assume is probably what they're planning to use it for, too. So, you can see some examples right here. And then you can plug in image and text. The agent thinks and observes it. And then you get the output, right? So, it can like zoom in and expect things. Uh, as you can see right here, you've got image annotation, so you can actually annotate the model, which we've got there. And then it's also got like visual maps and plotting as well. So, what are they planning to do next? Well, I think this is just like a really early technology, but basically, you know, one of the things will be like coddriven behaviors. So, it can zoom in on like smaller details, rotating images, performing visual maths, that sort of thing. Also, more tools. So I'm imagining that within the whole Google ecosystem, we'll be using aentic models more and more, right, based on what we need to do. And then you can see that you can actually get started today inside Google AI Studio and Vert. Ex AI using this code here to get started with it. Here's an example. So it can like look at a picture like this and be like, okay, here's workflow that we're going to do, right? Um and then it moves from this to this to this. This is a tweet by Omar um Sansa, as you can see. And uh yeah, it looks really cool. It looks quite interesting. I think it will run nicely within the ecosystem of Google. So, thanks so much for watching. If you like AI automation, feel free to check out the AI success lab link in the comments description. We've got loads of Google Gemini trainings inside there, including like how to use agents by Google Gemini. So, if you just type in Gemini on the search bar here, you'll see an archive and we've actually got a thousand lessons on Google Gemini, which is pretty insane. So like notebookm Gemini CLI, Google deep think uh Google agents etc. You can see for example we actually have a full course on like how to build and automate anything with Google Gemini and all these different lessons come with step-by-step instructions as well. On top of that if you haven't already check out the AI profit boardroom community. This is my AI automation community that shows you basically how to scale, save time and grow with AI automation. It's a community of people learning and growing and sharing what's learning right now. So, you can see here, for example, we got 2,200 members in there. We can actually connect you with people in your local area. So, depending on where you are in the world, and you see we've got people all over the world. Um, you can connect with people, zoom in on your local area, DM them, message them, ask them to jump on a Zoom call and meet up in real person, which is pretty cool. And then also inside here as well, we you can post questions inside the community here and get help and get support whenever you need it, right? So if you're struggling with anything, feel free to post inside here and we're always happy to help you. So for example, Radu posted yesterday about AI projects and you see all these people just getting back to him right here, including me personally. So I answer the questions personally inside there. You can also direct message me whenever you need it. And also you can jump on weekly coaching calls. You can see here we have live weekly coaching calls where you can get help, get support in real time. If you can't make the coaching calls, you can actually watch them back inside the community. Right here we have a fiveweek AI automation masterass that takes you from complete beginner to expert with AI plus how to build your first AI agent in under 5 minutes. If you like stuff like clawbot maltbot which is been trained in like crazy recently you can actually get a full 3-hour course on exactly how I set it up and my maltbot can for example like generate AI avatar videos. It can generate thumbnails. It can publish articles to WordPress and rank them on Google. Like it's insane what that can do. probably the most powerful technology of the year and you can actually use it for free. So feel free to check out the training on mobile and then we also have all the systems I personally use inside my business. So for example like how to automate your emails, X automations, newsletters, shorts, how we automate Instagram AI avatar videos etc. And then additionally you can see inside here we have all these trainings on like how to get more clients, how to rank with AI SEO, how to do YouTube and grow a YouTube channel with AI based on what's working for me.

And this is all inside the offer boardroom. So feel free to check out link in the comments description. Hope to see you inside there. Cheers.