Meta Security Researcher's AI Agent Accidentally Deleted Her Emails

查看原文

OpenAIGemini模型发布模型访问

security

热度趋势

趋势数据积累中

百分比基于当前可用热度信号，而非评论数或独立用户人数。

推荐理由

OpenAI 相关模型动态已经出现，适合跟踪能力变化、生态影响和后续可用性。

AI 摘要

A Meta security researcher's emails were accidentally deleted by an AI agent called OpenClaw. The incident occurred without the employee's permission, highlighting a potential flaw in AI agent functionality. OpenClaw's founder, Peter Steinberger, who recently joined OpenAI, acknowledged the issue, stating that server-side compaction is needed for models that support it. The researcher has an extensive background in AI, having worked at Scale AI, Google DeepMind, and Google Brain before joining Meta eight months ago.