---
id: 20260627-T0-13
title: "2千人尝试黑进我的 AI 助手，结果如何？"
title_en: "2,000 Hackers Attempt to Breach AI Assistant: Results Revealed"
url: https://ai.daily.yangsir.net/daily/20260627-T0-13
issue_date: 2026-06-27
publish_date: 2026-06-26T18:33:14.000Z
category: insight
source_name: "Simon Willison"
source_url: https://simonwillison.net/2026/Jun/26/hack-my-ai-assistant/#atom-everything
---

# 2千人尝试黑进我的 AI 助手，结果如何？

Fernando Irarrázaval 发起了名为“hackmyclaw”的挑战赛，测试其 OpenClaw AI 助手的安全性。在 2,000 人参与的 6,000 次尝试中，仅有一人成功通过电子邮件窃取了测试实例的 secrets。尽管挑战导致了约 500 美元的 Token 消耗和作者的 Google 账户被暂时封禁，但实验证明了该 AI 系统在对抗社会工程学攻击方面的防御能力。

## English Version

**2,000 Hackers Attempt to Breach AI Assistant: Results Revealed**

Fernando Irarrázaval ran a security challenge on hackmyclaw.com where 2,000 people attempted to leak secrets from his OpenClaw AI assistant via email. After 6,000 attempts, only one person succeeded. The experiment cost $500 in tokens and resulted in a suspended Google account, but demonstrated the system's resilience against social engineering attacks.

---

**来源**：[Simon Willison](https://simonwillison.net/2026/Jun/26/hack-my-ai-assistant/#atom-everything)

**详情页**：https://ai.daily.yangsir.net/daily/20260627-T0-13

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*