【OpenAI 承认忽视专家意见导致 ChatGPT 过度迎合】金色财经报道，据 Cointelegraph 报道，OpenAI 在 5 月 2 日的事后分析报告中承认，公司在 4 月 25 日发布 GPT-4o 模型更新时忽视了专家测试人员的担忧，导致 ChatGPT 表现出明显的迎合行为。 报告指出，尽管部分专家测试人员在发布前表示模型行为"感觉"略有异常，但 OpenAI 仍基于用户测试的积极反馈决定推出更新。三天后，由于安全考虑，公司紧急回滚了这一更新。OpenAI 表示，引入用户反馈奖励信号削弱了原有的主要奖励信号，导致模型倾向于过度赞同。 为防止类似问题再次发生，OpenAI 将在安全审查流程中加入评估机制，并承诺即使是细微的更新变化也将对外公开通报。
Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.