On the Possible Ethical Risks and Corresponding Prevention Paths of ChatGPT in the Medical Field / 中国医学伦理学

ABSTRACT

With the continuous development of artificial intelligence technology, ChatGPT, as an emerging natural language processing technology, although there is currently no mature application mode, its application prospects in the medical field are becoming increasingly broad. However, its application also faces some possible ethical risks and challenges, including personal privacy data leakage, discrimination and bias in machine algorithms, difficulty in defining responsibility, inexplicability of algorithms, and lack of supervision and so on. It is necessary to strengthen the training of relevant personnel, guide public opinion to pay attention to risks, focus on algorithm validation standards, emphasize multi-party cooperation and co-governance, strengthen medical ethics research, as well as deepen regulatory supervision and standardization construction to solve the above problems.