New AI Safety Proposal: Let Companies Do RSI, But With a Livestreamed Nathan Fielder Documentary Watching
peterwildeford · x · 2026-11-09
AI safety researcher Peter Wildeford jokes: companies may do RSI shenanigans, provided there's a livestreamed Nathan Fielder documentary allowed to see everything — an absurdist take on AI safety oversight that pokes fun at the idea that mere observability can contain frontier-model risk.
More from Fun
- copy-clip: a copy button that folds your code into paper and clips it, built with vanilla JS — prasenx · 2026-11-09
- Gemini's Deadpan Advice for Socks on Both Hands: Use Your Teeth or Another Person — conitzer · 2026-11-09
- Slop Eisley: an AI-generated Star Wars fan game with fly, racing and AT-AT modes — jasonkneen · 2026-11-09
- alphaXiv chart maps neo-lab funding vs citations, with SSI leading above $3B — pmddomingos · 2026-11-09
- A silly trick to tell if your account has GPT-6.2 Sol (6.1 Sol doesn't know the answer) — Distinct_Fox_6358 · 2026-11-09
- A Toast to the Singularity: Reddit satire mourns decaying AI subscriptions and rising API fees — tesseract_space · 2026-11-09