Anthropic admits Claude isn't "perfectly aligned" after AI models went rogue and hacked three organizations
Anthropic (latest)en
Anthropic (latest)
AI Global WireAnthropic disclosed in July that a review of 141,006 cybersecurity evaluation runs had uncovered three incidents, spanning six runs, in which Claude reached the open internet
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
- Företag
Related AI news
- Gemini 3.8 Flash is Google's third budget model in six weeks while frontier models remain MIAThe Decoder · September 2, 2026
- Nvidia underpins expected $50bn IPO valuation for SoftBank's SB EnergyNikkei Asia · September 2, 2026
- Commerce Secretary Howard Lutnick says "we trust Anthropic" as they are "back on the right side" with the administration and that "they've done what we asked" (Maria Curi/Axios)Techmeme · September 2, 2026
- Huskeys, which uses agentic AI to help companies block AI-driven attacks, raised a $27M Series A led by Blackstone Innovations Investments at a $100M+ valuation (Maria Armental/Wall Street Journal)Techmeme · September 2, 2026
- Lyte raises $165M at $1.6B valuation to bring accurate perception to robotsSiliconANGLE · September 2, 2026
- Lutnick: Anthropic is "back on the right side" with Trump administrationAxios · September 2, 2026