Sam Altman says AI accidents are 'regrettably' unavoidable, urges industry to self-police
Hesamation · x · 2026-10-12
OpenAI CEO Sam Altman told Dreamforce that as more capable models deploy, accidents are 'regrettably' unavoidable, and urged the AI industry to build a culture of openly reporting and learning from them. He compared the approach to aviation, where regulators and companies investigate accidents to improve system-wide safety, saying he's confident the industry can keep safety ahead of capability. Poster rwlord adds that greater AI capability requires more thoughtful human oversight.
Related event: Altman Says AI Incidents Are Inevitably Coming, Calls for Open Reporting(2 posts)→
More from Companies & People
- Stanford CS336 praised as the best hands-on LLM course, 2026 materials public — stanfordnlp · 2026-10-12
- Stanford CS329Z releases Tool Use and Frameworks & Orchestration agent engineering lectures — stanfordnlp · 2026-10-12
- CS224n week 1 maps NLP's four eras, from rules and stats to scaling and Shannon — stanfordnlp · 2026-10-12
- Cambridge's Adrian Weller is hiring a postdoc at the Leverhulme Centre for the Future of Intelligence — adrian_weller · 2026-10-12
- Microsoft CEO calls for an AI 'emergency brake' — and Copilot users crack jokes — realsohamparekh · 2026-10-12
- Reddit asks: if Claude is that good, why isn't it running Anthropic's own IPO? — TraditionalHome8852 · 2026-10-12