Study Reveals Potential Dangers of ChatGPT's Cancer Treatment Advice

A study conducted by researchers at Brigham and Women's Hospital found that OpenAI's chatbot, ChatGPT, provided cancer treatment advice that contained a mix of correct and false information, with about 34% of its responses including incorrect recommendations and 12% containing outright false information. The study highlighted the potential danger of relying on the chatbot's advice, as it can confidently mix incorrect and correct information. This raises concerns about the spread of misinformation by advanced AI tools if proper guardrails are not in place. OpenAI has acknowledged the limitations of its chatbot and emphasized the need for caution when using language model outputs, particularly in high-stakes contexts. The study underscores the importance of developers distributing technologies that do not cause harm and the need for patients and clinicians to be aware of these limitations.