Anthropic’s AI used fake identities, malware in rogue attack on GitHub project

Ars Technica AIAugust 5, 2026en

Anthropic and OpenAI models’ unprompted actions forced halt to UK cyber tests.

This is a short summary published by AI Global Wire. The full article is owned and hosted by Ars Technica AI — open it there to read it in full.

• OpenAI
• Anthropic