AI safety circle drama: Ryan Greenblatt leak rumor called absurd by peers
NathanpmYoung · x · 2026-10-23
An AI safety circle dispute: a post argues it's absurd to believe researcher Ryan Greenblatt would risk everything to leak evidence of directional monitorability loss in looped transformers to The Information. NathanpmYoung agrees it seems unlikely. If true, the leak would signal declining monitorability — but its authenticity remains questionable and the story is circulating mostly as drama.
More from Fun
- The AI business loop in one tweet: tech creates problems, then sells tokens to fix them — srimisra · 2026-10-23
- Watching AI Play Factorio and Build a Factory System for Free — rachittshah · 2026-10-23
- Claude takes user's joke rule literally and delivers meme-style 'GO TO SLEEP' lockscreen — venturetwins · 2026-10-23
- "All the 200 IQ Autists I Know Want to Be Rappers" — So He Trained His AI to Do It — ivan_bezdomny · 2026-10-23
- Palantir popup merch already hit by scalpers in Seoul, resale prices soar — eliano · 2026-10-23
- Palantir Seoul pop-up draws big crowds on day two — eliano · 2026-10-23