California AI report draws industry backing; proposals could become law in months

sjgadler · x · 2026-11-19

AI practitioner @w01fe publicly voiced strong support for multiple proposals in California's governor-led AI report and pledged to advocate for them. The quoted Sneha Revanur noted that RSI (recursive self-improvement) is mentioned near the top of the Governor's press release.

Notable ideas in the report include requiring companies to disclose what percentage of compute they spend on frontier AI R&D, creating outcome-based standards to complement safety cases, and pushing companies toward >SL3 model weight security. She argued that if California moves fast, parts of the report could become law within months, with lab employees' advocacy being a key determinant—calling it the #1 policy opportunity worldwide for managing RSI risks.

Original post →

More from Safety

Safety channel →