Silicon Valley is fully convinced AI agents represent the next era of tech. Inside the industry’s core, teams are building payment infrastructure for autonomous agents, leveraging these tools to automate internal work, and most recently, racing to block unregulated agents from hacking into outside organizations. But beyond the tech bubble, most ordinary people have never even interacted with an AI agent. That disconnect largely boils down to one thing: tech companies have not given the general public a good reason to care.

“This is a hot take… but isn’t it kind of crazy that almost no one actually uses AI agents?” Josh Miller, CEO of The Browser Company, wrote in a viral X post earlier this week. “Theoretically, the technology is ready for AI agents to totally transform how we work and live… but alas the general public dgaf [doesn’t give a fuck].”

Miller’s unfiltered take resonated widely across Silicon Valley, where many insiders remain genuinely bewildered that the public does not share their excitement for their latest obsession. Miller told me in a Wednesday interview that he dashed off the post while half-asleep during a family trip to Europe, but he stands fully behind his argument. He says the industry needs to shift its focus to building agent-powered products that people actually want to use. “I just have not heard a single person outside of the tech community mention an AI agent that they use regularly,” Miller says. “As excited and optimistic as we as an industry may be about the frontier and recursive self-improvement, it's worth pausing and just saying, ‘Huh, what are we actually building here?’”

Last month, OpenAI announced its Codex and ChatGPT Work agents draw a combined 10 million weekly active users. Sources close to Anthropic tell me the company’s Claude Code and Cowork agents see similar adoption levels. But compared to mainstream chatbots like ChatGPT and Gemini, which each average around 1 billion monthly active users, AI agents amount to little more than a rounding error in overall user counts.

That is a major problem for leading AI labs. Today, most consumers use generative AI for fairly simple tasks: looking up information, casual chatting, quick edits. But Silicon Valley has poured billions into training large models that can handle far more complex work, and AI agents are framed as the key to monetizing that investment—if consumers will actually use them. This lines up with broader concerns I’ve heard from AI insiders, who note that while the industry has poured massive resources into building AI agents, no one has launched a breakout consumer product for the space yet. The big, open question remains: when, if ever, will AI agents get their own ChatGPT moment of mainstream crossover?

Despite Miller’s self-deprecating demeanor—he frequently notes he is “just” a sociology major—I think his arguments are well worth paying attention to. His latest startup developed Arc, a web browser with a cult following, and was acquired by Atlassian last year for $610 million. His first company, Branch, was bought by Facebook in 2014, and Miller also previously served as the White House’s first director of product under President Obama.

Miller’s larger point is not just that AI agents have not caught on yet. It is that AI agents are a set of underlying technologies, not a standalone product consumers actually ask for. “No one wants AI agents, because AI agents aren’t a thing people are requesting. It is an invented frame made up by our industry to collectively refer to a set of capabilities,” says Miller. “Let's make a product that makes you calm, focused, and in flow right when you open your laptop. And the fact that we're able to do that because there's a thing called a ‘harness’ that calls up tools—who cares? No one needs to know that.”

Miller has firsthand experience to back up his claims. He says The Browser Company’s most popular feature ever is a personalized morning briefing for users in its AI-powered browser, Dia. When people open their laptop, they receive a homepage with a greeting, a daily to-do list populated from their calendar and email, and small, joyful touches like a curated piece of art. Technically, Miller says, the feature was only made possible by an AI agent, but the end user never needs to know that.

Miller’s arguments are convenient coming from someone who sells an AI-powered web browser, but I think he is correct that today’s AI agent products are largely not designed with mainstream consumers in mind. Tech companies are shipping the most impressive capabilities their models can pull off—such as navigating websites autonomously or writing full code bases—rather than creating finished products tailored to their customers' actual desires. Those capabilities may become great features someday, but right now, they feel more like tech demos than usable products.

Miller blames the disconnect on groupthink in the AI industry. Many people building these technologies are completely obsessed with them, and they often share a narrow, sci-fi-inspired vision for how the products should look and feel. Last summer, when The Browser Company was exploring acquisition, Miller says he met with many leaders at top AI labs. “It was wild to me. I think every single lab but one mentioned the movie Her as the way they articulated their vision,” says Miller, referencing Spike Jonze’s 2013 drama about a man who falls in love with an AI assistant. “I have a lot of respect for anyone that says, ‘Hey, here's where I think the world should go, and I'm going to go there.’ But I think there is sort of a lack of diversity of opinions and convictions.”

The latest trend in AI agents is giving users the power to build custom personal software to automate parts of their life, such as putting together pitch decks, organizing datasets, and other routine tasks. I’ve personally been using ChatGPT Work and Claude Cowork a lot, and I’ve found them to be far more useful than their predecessors. But there’s a limited pool of people interested in building their own productivity tools, and the industry needs far bigger ideas if it wants to reach the goal of mainstream adoption.

Miller says he hopes his message reaches beyond OpenAI and Anthropic and resonates with founders and product builders broadly. His goal is to push people to think outside the box about what AI agent products could look like, and then bring those new visions to life. “Let's not just accept this narrative of AI agents and actually question what are good products and tools that are joyful, useful, and approachable,” says Miller. “Who gives a shit if it looks like an AI agent? No one uses AI agents.”

This is an edition of Maxwell Zeff’s Model Behavior newsletter. Read previous newsletters here.