Technology news for families navigating serious mental illness — translated so you don’t have to.
We have an AI theme going on here. Recently, the focus was about AI talking to patients. This week is about AI writing about them — and it turns out that’s its own problem.
Telehealth is now the default way most people see a psychiatrist or therapist. Along with it came AI “scribes” — tools that listen to the visit and write the clinical note for the provider, so they can spend less time typing and more time looking at the person on screen. It’s a good idea. The execution isn’t quite there yet.
This week’s top story
Telehealth for mental health is now mainstream, but AI note-taking tools show major quality gaps
Medical Daily, citing a 2026 Medscape review — June 27, 2026
A new review of AI scribes used in telehealth visits found real gaps between what the AI wrote down and what actually happened in the room. Clinicians who’ve tested these tools say even the best ones still need a human to catch errors before the note goes in the chart. And right now, there’s no regulatory standard requiring that review to happen.
Here’s why that matters more than it might sound like it should: that note is what the next provider reads. It’s what gets pulled up at the ER. It’s what a psychiatrist skims before deciding whether to adjust a medication. If it’s slightly wrong — a missed detail, a misheard symptom — that error doesn’t just sit there quietly. It follows your loved one into their next appointment.
This isn’t a reason to panic about telehealth, and it isn’t a reason to assume every AI scribe is unsafe. It’s a reason to ask a simple question if you ever see a note that doesn’t sound right: “Did a person review this, or did the AI write it and nobody checked?” You’re allowed to ask that. You’re allowed to ask for a correction.
Meanwhile, check out Connected Care, where I cover AI in all the ways we need to understand in the world of serious mental illness. Its on Amazon!
Until next week,
Nicole
