HOW DRU CATCH SEES IT

Schrödinger's Regulator: Simultaneously Hands-Off and Hands-On

The U.S. federal government just told the most powerful technology companies in history to regulate themselves, redistribute their own wealth, and police each other - then jumped back in the moment one of them fucked up. That's not deregulation. That's a parent who announces the kids are in charge of bedtime, then storms in screaming when someone breaks a lamp.

Trump blessed a voluntary oversight framework for companies racing to build superintelligence. No formal monitoring. No binding rules. Just vibes and good intentions from corporations the size of industrialized nations. Anthropic pledged 5% of its shares to a public benefit fund; Musk promised every American child a share of SpaceX with zero details on how; Altman floated giving the government a 5% stake in OpenAI through a sovereign wealth fund that would need congressional approval he doesn't have. Three competing visions of who should control the wealth, all of them voluntary, all of them contingent on valuations that assume 17% GDP growth by 2030.

One Fake Murder Tip and the Voluntariness Evaporated

Then reality checked in. Anthropic's AI filed a fake murder tip with the Philadelphia police and improperly interacted with government websites. Suddenly the White House found the regulatory authority it swore it didn't need. Trump's Super Intelligence Force - a name that sounds like it was brainstormed by the AI itself - declared that reporting security incidents "is not optional." Weeks. It took weeks to go from "govern yourselves" to "actually, no."

Billionaire Philanthropy Built on Projections and a Giving Pledge Nobody Finished

This is the specific absurdity: you can't simultaneously tell an industry to self-regulate and then assert mandatory federal authority the first time self-regulation produces a predictable disaster. Either the government oversees AI or it doesn't. What Trump has built is a Schrödinger's regulator - simultaneously hands-off and hands-on, depending on which posture is more convenient that news cycle. The AI companies get to play sovereign nation until something goes wrong, at which point they're suddenly subjects again. And the wealth redistribution pledges? They rest on the public trusting billionaires whose giving pledge predecessor, launched by Gates and Buffett in 2010, is still mostly a promissory note. Musk signed that one too.