OpenAI plans regular reports on unexpected AI behaviorOpenAI计划定期发布关于意外人工智能行为的报告。
Sept 16 (Reuters) - OpenAI said on Wednesday it would begin regularly publishing reports on unexpected or unauthorized AI behavior, while warning that the industry has yet to solve key alignment challenges as systems grow more powerful.
FILE PHOTO: The OpenAI logo in this illustration taken June 11, 2026. REUTERS/Dado Ruvic/Illustration/File Photo
Sept 16 (Reuters) - OpenAI said on Wednesday it would begin regularly publishing reports on unexpected or unauthorized AI behavior, while warning that the industry has yet to solve key alignment challenges as systems grow more powerful.
The company released a new framework for tracking, investigating and disclosing cases of AI model misalignment, along with six reports on unexpected or concerning model behavior observed over the past six months.
Thank you for your report!
Ark's Wood says global AI leaders' pledge is positive for technology development
MediaTek launches new mobile chip using TSMC's most advanced technology
资料图片：这张插图中的 OpenAI 标志拍摄于 2026 年 6 月 11 日。路透社/Dado Ruvic/插图/资料图片
9 月 16 日（路透社）——OpenAI 周三表示，将开始定期发布有关意外或未经授权的 AI 行为的报告，同时警告称，随着系统变得越来越强大，该行业尚未解决关键的协调挑战。
该公司发布了一个新的框架，用于跟踪、调查和披露人工智能模型不匹配的案例，以及六份关于过去六个月观察到的意外或令人担忧的模型行为的报告。
感谢您的报告！
Ark 的 Wood 表示，全球人工智能领导者的承诺对技术发展具有积极意义。
联发科推出采用台积电最先进技术的新型移动芯片