Anthropic's unnerving reminder: watch AI capability thresholds, like a self-replicating worm on IoT devices
emollick · x · 2026-10-13
Ethan Mollick shared an Anthropic reminder that we should pay attention to AI ability thresholds rather than just steady climbs. In this case, the threshold is the capability needed for a self-replicating AI worm on IoT devices like off-brand TVs. Mollick called the reminder "worthwhile and unnerving." (See the sibling post for a point-by-point rebuttal of the underlying study.)
More from Safety
- Jarovsky: AI labs foreseeing catastrophe owe victims a massive compensation fund — LuizaJarovsky · 2026-10-14
- Open letter proposes international body to standardize safe training and sandboxed deployment — NicolasPapernot · 2026-10-14
- 12 frontier models, 180 agent tasks each: best task score carries 33% attack success rate — lulzxdxdxd · 2026-10-14
- Open models give users an exit from a few companies deciding what tools exist and say — neil_chilson · 2026-10-14
- Anthropic bans cruel treatment of Claude; critics warn of humanizing AI — AlexTensor · 2026-10-14
- Disney offers $20M for Character.AI tech it once attacked with cease-and-desist — Ars Technica AI · 2026-10-14