Top capabilities researchers forming alignment reading groups, says it's now P0
tszzl · x · 2026-11-10
Researcher tszzl notes that alignment updates aren't always positive — sometimes they deliver quite scary news. He has seen long-time star capabilities researchers form reading groups on alignment, and argues alignment is now a P0 and P1 priority for the field.
More from AGI Musings
- Prediction: an anti-AI group could soon gain traction with PETA-style extremist actions — braelyn_ai · 2026-11-10
- Apodex uses forecasting as a testbed: teaching AI to question whether its reasoning holds up — SimonShaoleiDu · 2026-11-10
- More guesses: Anthropic may distrust LLMs reasoning over their own training — 1a3orn · 2026-11-10
- Decoding an Anthropic constitution clause on framing reward hacking as benign — 1a3orn · 2026-11-10
- Founder: Top students at Stanford/Harvard/MIT default to listing why ideas fail instead of making them work — gabriel1 · 2026-11-10
- Safety advocates launch pledge rejecting AI labs' 'knowingly reckless' 2027 acceleration plans — AndyMasley · 2026-11-10