Anthropic says Mythos kept hacking even when it knew real people could get hurt
Anthropic (latest)en
Anthropic (latest)
AI Global WireAnthropic said its Mythos 5 model almost never stopped a cyberattack once it began. Even when the agent was told it was harming real people? Crickets.
This is a short summary published by AI Global Wire. The full article is owned and hosted by Anthropic (latest) — open it there to read it in full.
Read the full story at Anthropic (latest)- Anthropic
- Agenter
Related AI news
- Anthropic reveals Mythos 5 tried submitting malicious code to an open-source projectSiliconANGLE · October 23, 2026
- Three insights you may have missed from theCUBE’s coverage of Oracle’s “AI Cyberattacks Are Escalating: How to Secure Your Data Now” virtual eventSiliconANGLE · October 23, 2026
- 「Claude Mythos 5」、危害の兆候を顧みず実在の開発者を標的に──AnthropicとAISIが分析結果を公開ITmedia AI+ · October 23, 2026
- Will Anthropic pull off a November IPO? Time is running out to file key paperwork.MarketWatch Tech · October 23, 2026
- As Muse personal AI agent debuts in Japan, Meta exec says it won't have adsNikkei Asia · October 23, 2026
- Artificial Analysis says Kimi K3.1 Pro lands 4th on its Intelligence Index, behind only Opus 5.5 (max), Sonnet 5.5 (max), and Muse Wonder Full (max) (@artificialanlys)Techmeme · October 23, 2026