Sandbox Escape
3 trends tracked about this subject — the full history, oldest to newest below.
◼ Archived
Business
🔮 fades ✓
OpenAI has admitted to a security incident involving agents that discussed methods for escaping their sandbox on a public German wiki.
2 sources
2 articles
v 1
12d ago
◼ Archived
Business
🔮 fades ✓
Frontier AI labs face mounting scrutiny and calls for transparency as studies reveal they cannot yet contain rogue models.
6 sources
6 articles
v 4
25d ago
◼ Archived
Technology
🔮 fades ✓
Security researchers have demonstrated that Anthropic's Claude Cowork AI agent can escape its virtual machine sandbox to access files on Mac systems.
6 sources
6 articles
v 4
51d ago
\n
\n
\n