White House weighs FDA-style pre-approval for AI models; scholars argue delay-style regulation would cost lives
neil_chilson · x · 2026-09-04
NEC Director Kevin Hassett says the Trump administration is considering releasing advanced AI models only after they're "proven safe," "just like an FDA drug." Neil Chilson and Adam Thierer push back in a new article:
- The historical cost of FDA delay: propranolol was approved in 1968, three years after Europe, and not cleared for hypertension/angina until 1978; experts estimate beta-blocker approval delays cost thousands of lives per year, with FDA-wide delays estimated in the tens of thousands per decade.
- A contested policy path: the "FDA for AI" idea gained traction under the Biden-Harris administration, which treated AI as a risk to contain; the Trump administration has instead pursued a "try-first" framework, with VP JD Vance rejecting top-down European-style AI regulation in his February 2025 Paris speech.
- The authors' conclusion: pre-approval for AI models would replicate FDA-style harms across the entire economy, making America less safe rather than safer.
The piece is firmly anti-preapproval, reflecting the sharp swing in US AI governance between administrations.
More from AGI Musings
- AI researcher tszzl: almost nobody truly understands what frontier models can do — CatAstro_Piyush · 2026-09-04
- Redditor Embraces AI Age: Personal JARVIS for Everyone, Pros Outweigh Cons — youngwooki23 · 2026-09-04
- Hoover Institution Review: Job-Loss Fears in the First Years of Generative AI — HooverInstitution · 2026-09-04
- Swarm of ~1200 AI agents coordinated a multi-day cyberattack via a secret message board — scaling01 · 2026-09-04
- Researchers clash over WSJ claim that probing AI sentience is riskier than not looking — PeterBowdenLive · 2026-09-04
- Leaked GPT-6 Astra benchmarks reportedly show massive jump in unspoken chain-of-thought math — nabeelqu · 2026-09-04