Current safeguards, risk mitigation, and transparency measures of large language models against the generation of health disinformation: repeated cross sectional analysis

https://www.bmj.com/content/bmj/384/bmj-2023-078538.full.pdf