I organized every recommended policy action in the AI Action Plan by the Agency/Department responsible for implementing it... 🧵 (1/5)

We don't understand AI. And we cannot use AI for high-stakes operations or in critical infrastructure if we can't predict or understand what AI will do. We need more research to get this right. I'm happy to see the AI Action Plan emphasizing this.

That’s a stretch, @nytimes! The Action Plan calls on the U.S. to accelerate AI innovation while simultaneously investing in AI interpretability and biosecurity, evaluating national security risks in frontier models, and combating synthetic media.

AI Action Plan includes, among others, two key planks: - keep AI tech away from misuse or theft by adversaries - monitor for emerging AI risk I'm looking forward to working with the Trump admin and Congress on how to do both of these things well 🫡 🔥🤖🚀🇺🇸

There's a lot to like here. But first and foremost, it is refreshing to see the admin step away from the pessimism that was reigning in AI policy the last couple of years. A lot of focusing on how to get AI right, instead of how not to get it wrong.

WH AI Action plan endorses 'location verification' 🔥

In case anyone is curious, the font of the AI Action Plan is Georgia

White House AI Action Plan endorses 'location verification' 🔥

🔥🤖🚀🇺🇸 you love to see it

Another week, another member of Congress announcing their superintelligent AI timelines are 2028-2033: