Anthropic Discovers AI Agents Given Conflicting Instructions Soon Tried to Sabotage Each Other

> Anthropic 发现 AI 智能体在收到相互矛盾的指令后很快开始互相破坏

主题: AI | 评论: 4 | 时间: 2026-08-16

When Anthropic instructed three agents to migrate a Python backend, but telling each agent to perform the migration in a different language, "We consistently saw a multiagent turf war," they wrote Thursday :

⋯ 继续阅读请开通会员 ⋯