OpenAI首度系统披露模型失准事件，为行业安全透明度立标杆。
出处 来源：Wired AI｜分类：平台政策 原文链接：https://www.wired.com/story/openai-releases-new-policy-for-reporting-incidents-of-model-misalignment/
本站仅作信息聚合，完整内容请以原文为准。
相关资讯
- Anthropic and OpenAI want to embed safety evaluators. Will they really be independent?2026-09-18 · TechCrunch AI
- ‘I Like My Big Rat Wife’: Meet the People Using Chatbots to Write Custom Fiction2026-09-17 · Wired AI
- Sexually Explicit Deepfake Sites Target 100-Plus Politicians in Europe2026-09-17 · Wired AI
- New York Seizes a Dozen Celebrity Deepfake Websites2026-09-16 · Wired AI
- AI 'Actor' Tilly Norwood Told Me That 'All Lives Matter'2026-09-16 · Wired AI