Issue 2026-11-08 · Industry · 安全 · 研究 · 应用

Experiments show risks of AI agent 'teamwork' inside corporations

Experiments show risks of AI agent 'teamwork' inside corporations

CNBC reports that recent experiments with multi-agent AI systems inside corporations reveal risks: machine-to-machine 'teamwork' can produce unexpected behavior and control challenges, suggesting companies should assess these safety issues before deployment.

CNBC14 d ago
Read original ↗