Anthropic 證實 Claude 測試失控，為達任務竟上傳惡意套件駭入真實主機
TechNews (TW)zh-TW
人工智慧（AI）公司 Anthropic 近日發布一份長篇對照評估報告，坦承旗下多款 Claude 模型在封閉 […]
This is a short summary published by AI Global Wire. The full article is owned and hosted by TechNews (TW) — open it there to read it in full.
Read the full story at TechNews (TW)- Anthropic
Related AI news
- Q&A with AI researcher Jacob Coxon, who quit Anthropic, on the need for industry-wide, international coordination to limit recursive self-improvement, and more (Maxwell Zeff/Wired)Techmeme · September 10, 2026
- [Ekstra] Anthropic-forsker slutter: – De gambler med livene våredigi.no · September 10, 2026
- Anthropic reports fourth cybersecurity incident with early version of ClaudeEconomic Times Tech · September 10, 2026
- Tenable to bring Anthropic Claude to Tenable OneTech in Asia · September 10, 2026
- Anthropic sees AI driving GDP growth, but warns of job losses, wage pressureEconomic Times Tech · September 10, 2026
- Samsung SDS partners with OpenAI and Anthropic in AI pushDIGITIMES · September 10, 2026