Skip to Main Content

As an oncologist, Jennifer Lycette gets to know her patients particularly well. She’s doubtful that artificial intelligence could replace that personal connection, but new research based on, of all things, Reddit Q&As, says otherwise. The study, which Lycette wrote about in her essay, “Why a chatbot might seem more empathetic than a human physician,” found that chatbot-generated notes were, in some ways, better than notes from actual humans. The findings raised questions about the potential for using chatbots, like ChatGPT, to help providers answer patient questions submitted through electronic medical records — a task that can take hours of stolen, rushed time between appointments at the clinic.

“It’s not that we don’t like doing it, it’s that we’re just human, basically. But I don’t think that means we need to be replaced by any means,” Lycette said.


In this week’s episode of “First Opinion Podcast,” Lycette and host Torie Bosch discuss the potential for AI and ChatGPT in health care settings and what this technology could mean for patient and provider experiences in the future.

Be sure to sign up for the weekly “First Opinion Podcast” on Apple PodcastsStitcherGoogle Play, or wherever you get your podcasts.

And if you have any feedback for us — First Opinion authors to feature on the podcast, kudos, or darts — email us at [email protected] and please put “podcast” in the subject line.

Create a display name to comment

This name will appear with your comment

There was an error saving your display name. Please check and try again.