OpenAI and Anthropic Tell British Lawmakers: Regulate Us
There was something slightly surreal about watching Sam Altman and Dario Amodei sit before British lawmakers and essentially beg for a regulatory leash. A few hundred meters away, the head of MI5 was telling anyone who would listen that AI could help build new bioweapons. Two conversations about the same technology, happening at the same time, in the same city.
The shift in posture is worth paying attention to. Tech leaders have spent years treating regulation as a threat to be lobbied away. Now the two most prominent AI labs in the world are asking for oversight, and that tells you they've done the math. They know the window for shaping the rules is open now, and they'd rather be inside the room than outside complaining about what gets written. It's a strategic move, but that doesn't make it the wrong one.
I'd want to know what the actual proposals look like, though. Regulation is easy to ask for in principle and hard to get right in practice. The bioweapon warning from the spy chief is a useful reminder that this isn't just about abstract alignment research. The stakes are concrete, and the timeline is now. I just hope the rules they end up with protect everyone, not just the incumbents who asked for them.