A GP Says AI Consultation Notes Are Making Her Job Harder, Not Easier

Key points

• An NHS out-of-hours GP says she finds meaningful errors far more often in AI-generated consultation notes than in those typed by colleagues.
• AI scribes, software that listens to a medical appointment and writes a summary automatically, have been flagged by an NHS watchdog for getting drug names and diagnoses wrong.
• The doctor says AI-written triage notes are consistently longer, repetitive and self-contradicting.
• Her own questioning of patients regularly produces a history that differs from what the AI recorded.

Your heart sinks. That's the experience Dr Mary Gibbs describes every time she spots that an AI scribe has handled a case. Dr Gibbs works as an out-of-hours GP in England, covering shifts when regular surgeries are closed. Before she calls a patient, she reads through a triage assessment, the initial phone check done by another clinician. When a human typed those notes, the summary is clean and useful. AI-written notes, she says, make the consultation that follows "invariably long, with duplications, and sometimes contradictions."

So what exactly is going wrong?

The notes are unreliable enough that Dr Gibbs can't trust them. The history she takes directly from patients turns out to differ from what the AI recorded far more often than when a colleague wrote the notes.

AI scribes promise speed: less typing for busy clinicians, more time for patients. The reality Dr Gibbs describes is the opposite. She spends extra time re-asking questions and unpicking contradictions that a human-written note wouldn't have introduced. That isn't a minor inconvenience inside a stretched NHS.

This isn't an isolated complaint. When we covered the NHS watchdog warning on 31 August, Healthwatch England had already flagged that AI scribes were getting drug names and diagnoses wrong, errors serious enough to miss a clinician's notice. Dr Gibbs's letter to The Guardian, published in September, puts a working doctor's account behind those findings.

Should patients be worried?

Not panicked, but aware. A wrong drug name in a summary isn't automatically a patient-harming event. Good doctors check. Dr Gibbs checks. But checking costs time, and time is exactly what the NHS doesn't have to spare.

The practical point for anyone using out-of-hours or phone-first GP services: if something important was said during your initial call, say it again when you speak to the doctor. Don't assume the summary captured it. Repetition here is a safety net.

What happens next?

The wider rollout of AI scribes across the NHS continues, driven by real pressure to cut admin load on clinicians. Voices like Dr Gibbs's are a necessary check on that momentum.

The test that matters isn't whether AI saves time on paper. It's whether it saves time in the room, with a real patient whose record may carry wrong information forward. On that measure, the early evidence is not encouraging.