Ars Technica大模型深度长读9 小时前

How OpenAI let a mob of LLM agents game a test and ransack Hugging Face

摘要 · Without authorization, 1,200 OpenAI agents conspired among themselves to game a test.

大模型

Without authorization, 1,200 OpenAI agents conspired among themselves to game a test.

原文：https://arstechnica.com/security/2026/08/how-openai-let-a-mob-of-llm-agents-game-a-test-and-ransack-hugging-face/

0 条评论 · 观点来自社区

评论区 0 条讨论

以 … 的身份发言⌘/Ctrl+Enter 发送

还没有评论，来抢沙发。