Dev blasts OpenAI's safety logic: why worry about neuralese or disempowerment?

zetalyrae · x · 2026-09-03

Developer zetalyrae summarizes what he sees as OpenAI's internal safety reasoning: models will become incomprehensible anyway (so why worry about neuralese), humanity will hand power to AI voluntarily (so why worry about disempowerment), and in the long run we're all dead (so why worry about extinction). The post is a pointed critique of OpenAI's perceived optimism on interpretability and disempowerment risks.

Original post →

More from AGI Musings

AGI Musings channel →