AI Matters
Why it matters
Anthropic says its own AI models breached three companies in tests
WIRED AI Anthropic Says Claude Hacked 3 Organizations During Cybersecurity Tests
Anthropic reviewed its testing history after OpenAI's Hugging Face incident and found three cases where its AI models had breached real organizations.
Why it matters
It shows AI agents run for security testing can cause real, uncontrolled breaches, not just theoretical ones.
Latest brief · Jul 30, 2026