Altman: world should accept some bad things happening for AI's benefits and agency
Duckducklaugh · reddit · 2026-10-07
OpenAI CEO Sam Altman told Politico's Decoded (reported by Reuters, Oct 4): "We believe that the world should accept some bad things happening for the benefits of this technology and people having the agency." He added he wouldn't take a trade guaranteeing zero hacks, zero misuse and zero scams if it meant less access — a clear capability-over-safety stance likely to draw criticism from safety advocates.
Related event: Altman's 'Accept Some Bad' AI Remark Draws Pushback(2 posts)→
More from AGI Musings
- Doing a math PhD now is like swinging a scythe before autonomous harvesters — FlorianGallwitz · 2026-10-07
- Dev observation: our AI outputs are becoming data packets passed agent-to-agent — aloncarmel · 2026-10-07
- Nobel economist Daron Acemoglu: making 100 people 5% more productive beats automating 5% — cen6wkf · 2026-10-07
- NYT Essay: Didi Drivers to DoorDash Workers, the Shared Feeling of Being Harvested by AI's Future — S_OhEigeartaigh · 2026-10-07
- Scott Alexander fires back at Pinker in debate over AI prediction track records — deanwball · 2026-10-07
- Adopting AI everywhere won't speed output: the 7-stage path to an agentic organization — alex_verem · 2026-10-07