Three million people found OpenClaw in a matter of weeks. That number is the tell. Not because the product was polished — it wasn’t. It required a command line, a willingness to read instructions, and a certain tolerance for things going sideways in the background while you slept. Most people don’t have that tolerance. But three million people wanted it badly enough to try anyway. That’s not adoption. That’s demand with nowhere clean to go.
$GOOGL and Meta watched those three million and started sprinting. $GOOGL killed its earlier agent experiment and folded it into something called Remy — a round-the-clock assistant woven through Gmail, Calendar, and Search. Meta built Hatch in practice sandboxes, training it on DoorDash, Etsy, Reddit, letting it learn to navigate real consumer apps before it ever saw a real user. Both moves are rational. Both miss something.
The thing they’re competing to build isn’t an assistant. It’s a layer. The to-do list isn’t the product — it’s the interface they’re using to explain a product most people can’t yet picture. What they’re actually building is a permanent process running alongside your life, tracking your intentions, acting on your behalf, getting better at anticipating the gap between what you said you wanted and what you actually meant. That’s a fundamentally different thing than an app. It’s closer to infrastructure than software.
The Code Problem Underneath
Dario Amodei said something in a financial services event that was framed as a warning to SaaS companies and read by most as a threat. He said AI will make it cheaper to write software, the moat of complexity is dissolving, and companies that don’t adapt will go bankrupt. That’s the version that landed in the headlines.
The truer version is subtler. He said new moats — ones nobody can conceive of yet — may arise. That line got less attention.
He’s right. And here’s what that looks like in practice. When Anthropic’s coding tools launched, they sparked a selloff that erased $285 billion in tech stocks in 24 hours. The market read it as a destruction event. But the same forces that reduce the cost of writing software also reduce the barrier to building new software. If any sufficiently-funded team can now produce a working product in weeks instead of years, the question isn’t who survives the code deflation. It’s who figures out earliest what to build with all that cheap capacity.
The highlight from reading about vibe coding stayed with me: engineering teams are already choosing which language to write in based on what trains AI tools better, not what performs better in production. An organization considered migrating off Erlang — a language legendary for running resilient concurrent systems — because Java has more training data. The AI performs better, so the infrastructure bends to accommodate it. That’s a precedent that compounds. The tools don’t adapt to the codebase; the codebase adapts to the tools.
Robots at the Auto Show
Auto China 2026 was supposed to be about electric vehicles. The most interesting thing there wasn’t a car. It was the robots on the show floor — humanoid machines being demoed by companies that six months ago were known only for their battery supply chains or EV platforms. The vehicle manufacturers have decided that the same production logic that let them iterate on electric drivetrains fast enough to undercut legacy automakers applies to physical robotics too. High-volume manufacturing, fast iteration, cost compression. They already have the factories.
What’s worth watching here isn’t the robots themselves. It’s the transfer of industrial logic into a new category. The pattern that made EVs happen — where the incumbents dismissed the pace of improvement until it was too late to close the gap — may be running again, in humanoid robotics, on a faster clock.
That’s the through-line across all three signals today. Remy and Hatch are agents running in the background of your life. AI coding tools are agents running in the background of software production. Humanoid robots are agents running in the background of physical labor. In each case, the new thing isn’t replacing the old thing directly. It’s inserting itself into the process one layer back, making itself useful before it becomes indispensable, and then quietly becoming structural.
The companies watching OpenClaw reach three million users in weeks weren’t seeing a product. They were seeing proof that people will accept significant friction to get something running in the background that they trust. The friction will get solved. The trust question is the hard one — and it’s the only moat left that AI can’t write its way around.
Leave a Reply