被广泛用来进行事实核查的AI聊天机常常散布错误信息,研究发现。 AI chatbots widely used for fact-checking often spread misinformation, study finds.

像Grok,ChatGPT和Gemini这样的AI聊天机器人越来越多地用于即时的事实检查, 但研究表明它们经常传播错误信息. AI chatbots like Grok, ChatGPT, and Gemini are increasingly used for instant fact-checking, but studies show they often spread misinformation. 研究者发现这些机器人"一般不好" 拒绝他们无法准确回答的问题 反而提供不正确或投机的回答 Researchers found these bots are "generally bad" at declining questions they can't answer accurately, instead offering incorrect or speculative responses. 这一趋势的驱动力是技术公司减少人的事实核查员。 This trend is driven by tech companies reducing human fact-checkers. 这一转变引起了对机器人准确性和政治影响潜力的关切。 The shift raises concerns about the bots' accuracy and potential for political influence.