Gary Marcus criticizes OpenAI for dissolving three safety teams in two years
GaryMarcus · x · 2026-08-17
Gary Marcus cites reports that OpenAI has dissolved three AI safety teams in two years:
- Superalignment (May 2024): Co-founders Ilya Sutskever and Jan Leike resigned; Leike stated safety culture took a backseat to products.
- Mission Alignment (Feb 2026): Lasted 16 months before members were redistributed.
- Preparedness (End of July): The group evaluating catastrophic risks like bio-threats was dissolved.
Despite a pledge to dedicate 20% compute to safety, Marcus calls OpenAI the most disconcerting major AI company.
Related event: OpenAI Quietly Disbands Catastrophic Risk Team(9 posts)→
More from Companies & People
- Stripe to buy OpenRouter for $7B+; rival launches next day with 0% markup — kimmonismus · 2026-08-18
- Google DeepMind and Gradient Host Open Source Model Hackathon in SF — darian314 · 2026-08-18
- Chrome gets a new VP of Product as his startup tool Relay.app shuts down — laparisa · 2026-08-18
- Grok subscription bug offers $1000 deal with Cursor bonus — AGI Hunt · 2026-08-18
- Investing in the Machine Economy: Scarce Assets Beyond AI Generation — 0xSammy · 2026-08-17
- Poll: Which Cloud Provider Do 2024+ Startups Use? — yenkel · 2026-08-17