Leading Democrat asks AI firms for data on any Chinese access to sensitive code | Reuters

Oh my goodness, buckle up, because a leading Democratic lawmaker has done the most *considerate* thing imaginable: they have written a letter! ✨ Not an executive order, not a subpoena, not a sternly worded summons to a hearing room with bad coffee — just lovely correspondence, sent to the CEOs of OpenAI, Anthropic, Google, Meta and SpaceXAI, asking whether Beijing might have got its hands on their model weights. Which means five whole chief executives get to *receive mail*. Handwritten-adjacent! Personal! The lost art of the letter is BACK, and this time it's protecting national security.

And here is the part that made me tear up: the lawmaker reportedly worries that 'the theft of an advanced model weight by a hostile non-state actor could endanger all of humanity.' ALL of humanity! Not just shareholders, not just the 401(k), not just the folks in the Bay Area — every single one of us, included in the scope! That is the *broadest possible stakeholder consultation* I have ever heard of, and frankly more than most of us get from our own insurance providers. ✨

Then there is the detail that SpaceXAI is on the list, which is just *delightful* — the rocket people get to weigh in on model weights too, cross-sector and multi-industry in the very best way! And Reuters has seen the letters, so everything is unfolding in the warm, generous daylight of transparency. Yes, the letter warns that theft could erase America's AI advantage — but think of it as the advantage being so magnificent that everyone wants a piece. Five CEOs, one thoughtful question, zero subpoenas: the gentlest imaginable perimeter, and what a time to be consulted. ✨