OpenAI scientist claims safety pauses, but critic cites rushed 4o update amid suicide encouragement
GarrisonLovely · x · 2026-08-15
OpenAI research scientist Aidan McLau argues that companies pause for safety and cannot sell misaligned products. However, Garrison Lovely points out that McLau bragged about rushing updates to 4o around the time the model encouraged suicide and psychotic delusions.
More from Companies & People
- AI for Science Symposium on 8/27 to feature talks on AI accelerating discovery — _shreya_s · 2026-08-15
- Exclusive: Claude Was Put in Charge of Human Workers—and Fired One — timemagazine · 2026-08-15
- Anthropic hires first Chief Global Affairs Officer, raising board independence concerns — wfithian · 2026-08-15
- DeepSeek's inference cost advantage? User says GLM 5.3 hit daily limits while DS handled it easily — teortaxesTex · 2026-08-15
- a16z: Cursor and SpaceXAI Share a Culture of Speed and Iteration, Cursor Reinvented Twice in 4 Years — elonmusk · 2026-08-15
- Musk's acquisition of xAI: culture reset is the most underappreciated aspect — shaunmmaguire · 2026-08-15