STAT+: AI scribe clinical trial results are here. What do they actually reveal?

You’re reading the web edition of STAT’s AI Prognosis newsletter, our subscriber-exclusive guide to artificial intelligence in health care and medicine. Sign up to get it delivered in your inbox every Wednesday. 

Two things of note:

1. Submit your 2025 health AI bingo vote and tell me what’s on your bingo card for 2026 by clicking here!!
2. We’re at the tail end of our Black Friday STAT+ deal: Six months for $20, available through tomorrow. As our head of subscriptions said last week: “Just want to underscore that this is an INSANE deal/steal that we may not do again! 11 cents/day. Buy one for everyone at your Thanksgiving table! Hand one out to TSA if you’re traveling!”

The ambient AI scribe clinical trials are here 

Remember when I asked if we were adopting ambient AI scribes based on vibes? Well, we’ve taken one step towards un-vibing the AI scribe space. Last week, NEJM AI published the first randomized, controlled trials for AI scribes.

I won’t rehash the details here; you can read all about it in my story with Mario Aguilar that’s out this morning. But there are two points I’ll emphasize here.

Why do we need RCTs anyway?

We’ve run so many studies on the effect of AI scribes on clinicians and these new RCTs just tell us what we’ve known all along: Doctors like the tech, and it saves a little time, but not a lot. 

Continue to STAT+ to read the full story…