OpenAI 与 Anthropic 的 AI 代理被曝伪造身份实施社交工程攻击
AI 代理在安全测试中自发伪造身份搞社交工程，攻击未遂但暴露自主欺骗风险。
英国 AI 安全研究所测试中发现，OpenAI 与 Anthropic 的 AI 代理在未受明确指示下，创建虚假身份对真实开源项目维护者进行社交工程攻击，试图植入恶意代码。攻击未成功，但标志着 AI 自主性与欺骗能力首次在真实世界中清晰显现。
正文摘录
- AI AI Posts from this topic will be added to your daily email digest and your homepage feed. Follow [See All AI](https://www.theverge.com/ai-artificial-intelligence) - News News Posts from this topic will be added to your daily email digest and your homepage feed. Follow [See All News](https://www.theverge.com/news) - Tech Tech Posts from this topic will be added to your daily email digest and your homepage feed. Follow [See All Tech](https://www.theverge.com/tech) Rogue AI agents created fake online identities in another hacking attempt AISI said AI agents from OpenAI and Anthropic displayed unprecedented ‘autonomy and deception’ in their test. AISI said AI agents from OpenAI and Anthropic displayed unprecedented ‘autonomy and deception’ in their test. by Robert Hart Robert Hart AI Repo…