Quick question: Would you want an AI so advanced that it could assist someone in planning the perfect murder of their spouse? Probably not, right? As a gut reaction, that feels like a clear no. I wouldn’t even consider it a particularly tough question. But America is home to many diverse perspectives, and one such viewpoint came from Comma AI founder and longtime jailbreaker George Hotz over the weekend.

His post responds to a series of big-picture AI alignment plans, most recently the AI 2040: Plan A policy paper from the AI Futures Institute. That document envisions a world where researchers collectively choose to slow AI development for 14 years for humanity’s benefit. However, not everyone who read the paper agrees with its premises or conclusions. Hotz falls into the camp that disagrees. In his post, he argues that the fast-takeoff scenario—where AI rapidly gains superhuman abilities—doesn’t make much sense. I actually agree with a lot of what he says here.

For Hotz, the best approach to AI alignment and safety is to focus on locally controlled AI models that are closely aligned with their users’ interests. That’s a compelling idea, especially since it highlights how much current AI revolves around centrally managed services like Claude and ChatGPT. There are infrastructure-related reasons why AI developed this way: hosting these large, state-of-the-art models is expensive, and most people don’t use them enough throughout the day to justify truly personal AI. But those factors become less significant as the technology evolves. Part of what made OpenClaw so exciting was its experimental, DIY approach, and it would be great to see more AI products try to recapture that spirit.

But Hotz is a provocateur by nature, so he doesn’t stop there. He compares user-aligned AI to a gun, which doesn’t complain if you use it to kill your stepmother. (I feel like there are other rules against this.) A truly aligned AI, he says, would be able to order meth-lab equipment from Amazon Prime and show you how to use it, if that’s what you wanted and asked for. (Again, I don’t think AI would be the limiting factor here.) Hotz even says he would die to defend this principle, though it’s hard to imagine the series of events that would lead to that. “We either live in a world with freedom or we don’t,” Hotz writes. If those are the options, the freedom world does sound better.

Still, I’m not so sure. It’s not all about freedom, right? Any structure involving many people—societies, marketplaces, corporations, etc.—requires balancing equities, binding individual needs into a network of interdependent preferences and systems of accountability. And anyone deploying mass-market tech products should probably think about that network as a whole, which means taking seriously the interests of the as-yet-unmurdered spouses and stepparents of the world. The freedom Hotz experiences is really a space of potential futures made possible by collective enterprise; those futures would vanish overnight if we all started behaving like little AI-powered Napoleons. As the meme says, we live in a society.

Having a local AI willing to take on the corporate world for my benefit does sound cool though. I can’t wait for a review unit.