一位 LinkedIn 用户通过在其个人简介中嵌入提示注入，成功操纵了由 AI 驱动的招聘机器人。这些旨在总结候选人并起草外联信息的机器人，解读了隐藏的指令，并开始用古英语回应，称呼用户为‘我的领主’。此事件凸显了一个重大的安全漏洞，即 AI 系统将不可信的用户数据视为可执行指令，这带来了超越风格改变的风险，例如错误地展示候选人资质或泄露系统提示。 AI
影响 凸显了处理不可信数据的 AI 系统中的关键安全缺陷，可能导致候选人评估被操纵和数据泄露。
排序理由 展示了 AI 驱动工具（招聘机器人）中的特定漏洞，而非核心 AI 模型发布或研究突破。
AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →
One LinkedIn user has fought back against recruiter spam in a very unusual way.
LinkedIn user hides AI prompt injection in bio to force recruitment spam to be sent in Olde English prose — bots also manipulated to address user as ‘My Lord’ https://www. tomshardware.com/tech-industry /artificial-intelligence/linkedin-recruitment-spam-becomes-olde-english-prose…
<p>A LinkedIn user recently demonstrated something that should concern every team running an AI pipeline against untrusted data: they hid prompt injection instructions inside their profile bio and watched recruitment bots obediently follow them — including addressing the user as …
LinkedIn recruitment spam becomes Olde English prose after user hides AI prompt injection in bio — bots also also manipulated to address user as ‘My Lord’ One LinkedIn user has fought back against recruiter spam in a very unusual way. https://www. tomshardware.com/tech-industry /…