
Can AI Be Harmful? A Conversation with MIT’s Dr. Marzyeh Ghassemi
NEJM AI Grand Rounds
How to Trust a Black Box AI Tool
The question that I hear a lot from clinicians primarily is about trust. And they will say something like, how can I use this black box AI tool if it can't explain it's reasoning to me? We have very, very good evidence that in human decision making models with more perceived transparency or explainability hamper people's ability to detect when a model is making a serious mistake. The same people are able to detect those mistakes in quote unquote black boxes where there is no explanation.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.