Challenges for Ethics Review Committees in Regulating Medical Artificial Intelligence Research

The integration of artificial intelligence (AI) into medical research poses significant challenges for Ethics Review Committees (ERCs), especially concerning clinical trials and ethical assessments. This review identifies 23 primary challenges ERCs encounter in regulating medical AI, categorized into five sections that explore limitations faced by ERCs, researchers, and regulatory bodies, along with the complexities of rapidly advancing AI technologies. Key issues include the intricacies of AI systems, risks related to personal data re-identification, and the impact of multi-subject data on algorithm outputs. Ethical concerns about algorithmic bias, data ownership, and AI commercialization are also highlighted. The variation in regulatory frameworks across regions further complicates these challenges, highlighting the urgent need for revised guidelines tailored to AI research. The article stresses the necessity for interdisciplinary collaboration and ethics education for all stakeholders in AI, advocating for continuous reviews and enhanced engagement to ensure informed consent and address potential unintended consequences. Ultimately, it calls for the adaptation of ethical oversight mechanisms to support innovation while upholding ethical standards in the quickly evolving field of medical AI.