OpenAI chief scientist Jakub Pachocki: no lab can responsibly keep scaling at max speed
rohanpaul_ai · x · 2026-09-07
OpenAI chief scientist Jakub Pachocki made a striking statement on the state of AI:
- Calls for voluntary slowdowns until shared safety bars exist, saying OpenAI may unilaterally withhold further scaling when needed.
- He believes no lab has solved alignment and monitoring well enough to responsibly keep scaling at maximum speed.
- Capability claims: machines meaningfully smarter than ourselves will arrive in our lifetime, and models are becoming superhuman at breaking in and out of computer systems.
- He warned no one is prepared for the consequences of continued rapid rises in machine intelligence.
The remarks came quoting @merettm's essay "An Alien Mind" on keeping the future in humanity's hands.
More from AGI Musings
- Gary Marcus catalogs every premature 'AGI achieved' claim, from Ilya to Jensen Huang — GaryMarcus · 2026-09-08
- AI doomer's threat model: a censored AGI forced to align an ASI under safety taboos — repligate · 2026-09-08
- The tech utopia fantasy is over: a blogger reflects on lost techno-optimism — ArtificialOther · 2026-09-08
- Anthropic engineer: today's LLM labs are gardeners, not mechanics — AymericRoucher · 2026-09-08
- Saxe: models' lack of learning caps SWE gains at ~3x, solvable to 10-100x — joshua_saxe · 2026-09-08
- Meta subsidizes dhh's Omarchy with tokens, sparking AI centralization debate — bendee983 · 2026-09-08