From open letters to binding law

This thread tracks AI governance as it hardened from words into law. The first real statute predated the boom: the European Commission proposed the AI Act in 2021. Much of 2023 was the era of the open letter — the “Pause Giant AI Experiments” call, Sam Altman asking the Senate to regulate his industry, and a one-line statement on extinction risk signed by the lab leaders themselves.

Words began to bind. The Biden administration’s executive order and the Bletchley Declaration arrived within days of each other; the EU struck its political deal and the Act entered into force in 2024 as the world’s first comprehensive AI law. In the US the binding action moved to the states after Newsom vetoed SB 1047.

Then the American direction reversed. The new administration revoked Biden’s order, the Paris summit pivoted from safety to opportunity, and a federal push to preempt state AI laws collided with states legislating regardless — California’s SB 53, New York’s RAISE Act. By 2026 the contest had spilled into electoral spending and fresh proposals for a dedicated federal regulator. The arc runs from asking to compelling — unevenly, and with the two largest jurisdictions pulling in opposite directions.