Shanghai,China-Nov.22nd 2024:Claude, Gemini and ChatGPT AI app icon on screengettyOne of the hallmarks of the early days of 2025 is this rapid-fire news coming from the AI industry. Models are evolving, week after week. Companies are throwing enormous amounts of money at everything from hardware to specialized use cases. Some of the fruits of this labor are now coming out as consumer technology, too.One very good example is the two-fold process over at Google. Right now, Gemini 1 is getting put into endpoint devices, and introduced to the average user. But Gemini 2 is also underway in a big way, and people close to the project are coming around with demos of the new AI agent technology.And then theres Googles Notebook LM. Its turning heads because through its quick evolution, Notebook LM is now much more than just a note-taking tool. Google is showcasing new Notebook LM capabilities. I saw one of these demos recently, and it was really illuminating. Google is embracing the idea of AI agents, and showing the world how this kind of thing works.Agents and TasksIf you look at the Google blog where Sundar Pichai and others are explaining the building of this agentic technology, theyre talking about native tool use. That means that AI entities are going to be able to use tools in the same ways as humans do.MORE FOR YOUWere getting 2.0 into the hands of developers and trusted testers today, the leadership team writes in an announcement that will likely have a big impact on the industry. And were working quickly to get it into our products, leading with Gemini and Search. Starting today our Gemini 2.0 Flash experimental model will be available to all Gemini users. We're also launching a new feature called Deep Research, which uses advanced reasoning and long context capabilities to act as a research assistant, exploring complex topics and compiling reports on your behalf. It's available in Gemini Advanced today.So the AI agents are coming. What that means, in in turn, is that theyre going to start doing the tasks that have always been assigned to people.The DemoLooking at an actual demo of Notebook LM recently, I saw how a sales person, for example, could automate various parts of their day. The AI agent can contact the person who was the last touch point in the sales funnel, and get information from them. The AI agent can create a roadmap to opportunities. The AI agent can go through a proprietary content management system, and look at everything thats been posted in the last two weeks, and make decisions accordingly.The sky is the limit.And it raises questions about what our roles are as people in a business processAgents and PlanningHeres another part of that demo that caught my attention. Its where the presenter used Notebook LM to come up with long-term planning results. Suppose you want a five-year plan for a project rollout. Or you just want some swot analysis. Prior to, lets say, 2021, people would put up Post-it notes with various parts of the process, and arrange them to get their results. Now we can just let the AI agents do it. We dont have to ask them how its done, necessarily - just sit back and let them do their work.I also came across this interview with Geoff Hinton with him explaining how agentic AI could likely take over a lot of control from its human handlers. Thats sort of the other side of the coin to think about, as we see these technologies evolving by leaps and bounds.And then theres search.No product has been transformed more by AI than Search, Sundar Pichai writes. Our AI Overviews now reach 1 billion people, enabling them to ask entirely new types of questions quickly becoming one of our most popular Search features ever. As a next step, were bringing the advanced reasoning capabilities of Gemini 2.0 to AI Overviews to tackle more complex topics and multi-step questions, including advanced math equations, multimodal queries and coding. We started limited testing this week and will be rolling it out more broadly early next year. And well continue to bring AI Overviews to more countries and languages over the next year.All of that is pretty exciting.Taking Stock of Your AssetsThere was one other use case that I mentioned personally, because I think that its probably going to be more popular than most others. That relates to the use of Gmail.Our inboxes are overflowing with content. Theres so much that we have trouble keeping up sometimes, at least, I know I do. What if the AI agent could simply scrape through your Gmail account and bring you whats most relevant and most actionable?That would be one of those blue star applications for a technology that, in my estimation, fits the bill for its zeitgeist. So lets keep an eye on this emerging product, because it just might be one of the primary catalysts of AI agent use in our time.