UK AISI 发布首篇论文：用审计对抗游戏检测模型故意藏拙
AI 导读
UK AISI Model Transparency 团队发布首篇论文，研究能否识别刻意隐藏自身能力的 AI 模型。团队搭建了一个审计对抗游戏：红队构造存在条件性表现不佳（故意藏拙）的模型，蓝队尝试将其识别出来，结果显示红队获胜。团队成员 Thomas Read 表示，团队构建了一批条件性表现不佳的模型生物，并测试了多种检测技术在对抗环境下哪些有效。
正文
First paper from my team at UK AISI! Excited to have this out there - we have some really great model organisms of conditional underperformance, and tried a lot of different detection techniques to see what works in an adversarial setting
NEW PAPER from UK AISI Model Transparency team:
Could we catch AI models that hide their capabilities?
We ran an auditing game to find out. The red team built sandbagging models. The blue team tried to catch them.
The red team won. Why? 🧵1/17在 X 查看被引用的帖子