June 17 2026 - AI Market Shakeup: Gemini Surges & New Agent ToolsAI Market Shakeup: Gemini Surges & New Agent Tools

Hey everyone, welcome back to another episode of AI Signal & Noise. It is Wednesday and dude, we have some absolutely wild stories to unpack today. I am Taylor. And I am Morgan. We're looking at everything from shifting market shares to some pretty deep tech standards and agent workflows today. Ready to get into it, Taylor? Oh, absolutely ready. Let's dive straight into our first story because honestly, this one blew my mind a little bit. So, I saw on TechCrunch that ChatGPT's market share has officially slipped below 50% for the first time ever. Like, they are still the biggest, but the crown is slipping, dude. Wait, really? Below 50%? That is actually huge. I mean, they basically define the space. Who is eating their lunch? Is it Google? You called it. Gemini is right behind them with 662 million monthly users. ChatGPT still has over 1.1 billion, but the gap is closing fast. Wow, 662 million for Gemini? That is massive. Though, to be fair, Google pushes Gemini onto every Android device and browser, so those numbers might be slightly inflated. Totally, but still. And guess who is third? Claude is sitting at 245 million monthly users. People are really loving Anthrox lately. Honestly, that makes sense. Claude has been a developer favorite for coding and reasoning, but 50% is a psychological barrier. OpenAI was to be feeling the heat. Oh, for sure. They are not the only game in town anymore. The competition is getting so fierce and users are realizing they have actual solid options. It is a good thing for us, though. More competition means these companies have to keep innovating and hopefully lower their prices. Exactly. It is a massive win for consumers. But speaking of Google and how they are trying to organize all this AI stuff… Right. They are not just pushing Gemini. They are trying to build the underlying infrastructure for how these models actually talk to our data. Yes. Which brings us to this super cool new open standard Google Cloud just dropped. It's called the Open Knowledge Format, or OKF. Ah, yes. OKF. I read about this on MarkTech Post. It is essentially an open, vendor-neutral markdown specification designed to give AI agents curated context. Dude, it sounds so simple, but it's actually genius. It formalizes what they call the LLM wiki pattern. Basically, a directory of markdown files with YAML front matter. It is highly practical. Instead of relying purely on vector databases and complex REG pipelines, OKF allows developers to explicitly structure knowledge. Each concept just needs a type field. Right. And because it's markdown, humans can easily read and edit it, but AI agents can parse it perfectly. It's like the ultimate middle ground, you know? Exactly. They focused on three design principles. Simplicity, portability, and predictability. It is vendor-neutral, so you are not locked into Google's ecosystem at all. That is so cool of Google to make it open source. They even shipped reference tools like a Python consumer and an interactive bundle explorer you can embed. But let's be realistic here. Will developers actually adopt this, or is it just another standard destined to be ignored? We already have so many ways to format data. I think they will. RAG can be so messy and unpredictable. Having a clean, curated markdown bundle just makes sense for complex agent workflows. True. Deterministic context is always better than fuzzy search when accuracy matters. It is definitely a step towards more reliable AI agents. Speaking of reliable and, like, seriously powerful AI agents, NOS Research just did something wild with their HERMES agent. Oh, the HERMES update? That is a big deal for anyone building complex agent networks. They're tackling the bottleneck problem. Exactly! They just added asynchronous sub-agents. Dude, this is a game-changer for how these agents multitask. Right. So normally, when a parent agent delegates a task to a sub-agent, the whole chat blocks. You just sit there waiting for it to finish. Which is so annoying, right? It's like waiting for your co-worker to finish an email before you can even type anything else. Exactly! But with this new update tracked under issue 5586, the delegate tool can spawn background sub-agents that don't block the parent chat at all. Yes! You can spawn them, check their progress, steer them if they get off track, and collect the tasks when they are done, all while chatting. It introduces sub-agent isolation too. But we have to consider the trade-offs. Async workflows are notoriously harder to debug and manage than synchronous ones. Oh, totally! If a sub-agent goes rogue in the background, you might not notice until it has wasted a bunch of API credits, right? Precisely. There is a reason they discuss when to use this versus ACP, or Agent Communication Protocol. It is not a silver bullet for every task. But for heavy multitasking? Like, if you want an agent to research five different topics at once while you keep brainstorming, that is incredibly powerful. Definitely! It makes the AI feel much more like an actual team of assistants working in parallel, rather than a single slow chatbot. I love that! And you know what else is making AI feel less like a chatbot and more like a full-blown software team? This new tool called Atoms. Ah, the vibe coding tool. I have been seeing this term everywhere lately. It sounds a bit like a marketing buzzword, honestly. Haha, maybe. But the concept is so cool. You literally just describe your idea in plain language and it does everything for you. Let's break down what everything actually means here. Because building an app is one thing, but deploying and marketing it is another. Well, that is the crazy part. Atoms doesn't just write the code. It uses AI agents to build, deploy, and market your app. No code required. Wait, market it too? How does an AI agent market an app? Is it just spamming social media or is it doing actual target analysis? From what I gather, it helps set up landing pages, drafts promotional content, and optimizes SEO. It's like having a tiny startup team in your browser. Okay, that is interesting. But as a skeptic, I have to ask, what happens when the app breaks? If you don't know how to code, how do you fix it? You just vibe-coded. You tell Atoms, hey, this button isn't working, and the agents go in, find the bug, and deploy the fix automatically. I mean, that sounds amazing in theory. But real-world software architecture is complex. I worry this will lead to a flood of poorly built, unmaintainable apps. Sure, but it also lowers the barrier to entry. Think of all the creators and entrepreneurs who have great ideas but zero coding skills. This empowers them. That is a fair point. Democratizing creation is great. I just hope the underlying code quality is solid enough to handle actual scale. Only time will tell, but honestly, the fact that we are even talking about AI agents marketing our apps is wild. What a time to be alive. It certainly is. From market share shifts to asynchronous coding agents, the landscape is moving incredibly fast. It really is. And that is a wrap on our main stories for this Wednesday. Well, we covered a lot of ground. It is clear that the tools are getting more sophisticated and the competition is heating up. Absolutely. It is going to be fascinating to see how these new standards and tools shape the next wave of AI apps. Definitely. Thanks for joining us today, everyone. Make sure to subscribe and check out the show notes for all the links. Yes, definitely check those out. That is it for today. We will be back tomorrow with more AI news.