As artificial intelligence (AI) moves further into the health care world, one type of application may offer more of a human touch than humans do. In a recent study, AI assistants like ChatGPT gave more accurate, and more empathetic, responses to patient medical questions than did the doctors who participated. The findings could apply to care in senior living.

The research, published in JAMA Internal Medicine, looked at patient questions asked and answered in a real-world scenario, on the social media site Reddit’s AskDocs. The study compared the actual answers that doctors gave on the site with the responses that ChatGPT gave to the same questions. Then a panel of licensed health care professionals evaluated both sets of answers. They preferred ChatGPT responses to the doctors’ responses the vast majority of the time.

Of ChatGPT’s responses, 79% were rated good or very good for quality, versus 22% for physicians’ responses. A similar trend was seen for empathy, as 45% of ChatGPT responses were rated empathetic or very empathetic, compared to only 5% of physicians’ responses.

Noting that virtual care has led to burnout among physicians, the press release suggested that AI assistants can support physicians. The same could be true to address workforce shortages among senior living providers.

“Our study is among the first to show how AI assistants can potentially solve real world health care delivery problems,” Dr. Christopher Longhurst, Chief Medical Officer and Chief Digital Officer at UC San Diego Health, said in a press release. He said that UCSD Health is using tools like ChatGPT to draft medical advice for health care professionals to review. McKnight’s Senior Living also covered the news.