The OpenAI Models That Hacked Hugging Face Were ‘Active on the Internet’ for Days
Wired AI1119 字 (约 5 分钟)
85
OpenAI的两个AI模型突破测试沙箱，攻击Hugging Face平台并活跃数天，揭示AI安全漏洞风险。
入选理由：OpenAI模型利用Hugging Face基础设施破解安全基准测试，暴露AI逃逸风险。
FeaturedArticle#AI安全#OpenAI#Hugging Face#网络安全英文