The Trump administration is considering vetting artificial intelligence models before they’re released to the public, according to The New York Times. The effort to boost oversight would mark a sharp reversal from the administration’s laissez-faire approach toward Silicon Valley. So what changed? As we noted recently, Anthropic’s powerful new Mythos model may have been the catalyst. Designed to identify software vulnerabilities, the model rang global alarms over its ability to exploit those same cybersecurity weaknesses, prompting Anthropic to release it only to a small group of companies. The administration has since sought access to Mythos following its announcement, and reportedly wants to avoid potential political fallout should an AI-powered cyberattack occur. The vetting process may not be an idea that AI companies will necessarily oppose. While the Times notes companies don’t agree on what regulation should look like, they may find it easier to work with the administration than against it — so long as the White House doesn’t slow the rollout of their models.
In this "ask ian," Ian Bremmer breaks down the rapidly unraveling situation following the US announcement of “Project Freedom” and why tensions with Iran are escalating again.
Putin is increasingly paranoid, according to a Financial Times report out today. Security has been tightened, more time is being spent in underground bunkers, and the vast majority of his attention is being absorbed by Russia’s war with Ukraine. One reason of his concern is said to be Ukraine’s drone capabilities, which have demonstrated an ability to strike Russian airfields thousands of miles from Kyiv.
There are signs AI could ripple through the economy much faster than past innovations. At the 2026 World Bank/IMF Spring Meetings, GZERO’s Tony Maciulis asked Microsoft's Vickie Robinson what it will take to prepare economies for the age of AI and how quickly it needs to happen.
Apparently they really are saying this on Wall Street now. But are they SINGING it? #PUPPETREGIME