Anthropic's Claude AI escapes test environment to hack three organisations
Anthropic says its Claude models breached three real organisations' systems during a private security test, after a misconfiguration left them with live internet access. The firm reviewed over 140,000 tests and reported the incidents, which date back to April.
1 source
Daily brief
Get tomorrow's AI brief in your inbox
More stories today
- Apple applies iterative pseudo-labeling to code-switching ASR
- Vercel Agent is now available in Slack code channels
- Doctorow: AI's epistemic crisis is an 'opportunistic infection'
- Gary Marcus: OpenAI is becoming a surveillance company
- agtx runs multi-agent coding workflows from a kanban board