PerplexityがAIエージェントの“暴走”対策ツールをオープンソースに Claude CodeやCodexを監視
ITmedia AI+ja
ITmedia AI+
AI Global WirePerplexityがAIエージェントの危険な挙動を検知・防止するツール群「Numbat」をオープンソース化。Claude CodeやCodexに組み込み、タスクに執着したエージェントの“暴走”を実行前に阻止する。
This is a short summary published by AI Global Wire. The full article is owned and hosted by ITmedia AI+ — open it there to read it in full.
Read the full story at ITmedia AI+- Anthropic
- Sök-AI
Related AI news
- Meta, Anthropic, Google, OpenAI to meet Trump officials about AI safety testingEconomic Times Tech · August 4, 2026
- 이파피루스, 열람과 비식별화를 하나로! '스트림닥스 7.0' 신제품 출시ETNews (KR) · August 4, 2026
- White House invites AI companies to review its new AI safety frameworkSiliconANGLE · August 3, 2026
- Big Tech's Anthropic and OpenAI stakes are distorting the corporate earnings pictureCNBC Technology · August 3, 2026
- Alibaba's new Qwen model is also taking your job, but this time it's greatThe Decoder · August 3, 2026
- Anthropic's Claude to infer, process data locally on Indian servers for local customersEconomic Times Tech · August 3, 2026