Study: 30% of UK GPs use AI tools in patient consultations

Artificial intelligence has slipped quietly into the exam rooms of Britain. What once felt taboo now lives on the screens of family doctors. A new study shows that almost three in ten GPs in the UK lean on AI tools such as ChatGPT during patient consultations. They do it to cut through paperwork, to sharpen clinical thinking, and, sometimes, to simply catch their breath in a job that feels more relentless by the week. Yet this rise comes with a nervous edge.

As one researcher put it, there is a "wild west" feel to the whole thing, since the rules that should guide doctors have not kept pace.

How does it work?

GPs are turning to AI in simple yet time-saving ways. The survey by the Royal College of GPs and the Nuffield Trust found that many doctors use AI to tidy up their day.

They rely on it to:

• draft clean summaries of patient appointments
• support diagnosis of common conditions
• speed up routine office tasks

A total of 598 out of 2108 GPs said they already use AI in some form. Men use it more than women, and usage is far higher in affluent areas than in struggling ones. Some NHS integrated care boards actively encourage it, while others forbid it, creating patchy, confusing rules.

As Dr Becks Fisher put it, "The government is pinning its hopes on the potential of AI to transform the NHS. But there is a huge chasm between policy ambitions and the current disorganised reality of how AI is being rolled out and used in general practice."

Why does it matter?

This shift matters because it shows how quickly tech can change the rhythm of care. In just one year, usage rose from 20% to 25%, according to a separate Digital Health study. The lead author, Dr Charlotte Blease, captured the speed of it perfectly when she said, "In just 12 months, generative AI has gone from taboo to tool in British medicine."

But the stakes are real:

• Doctors fear legal trouble if AI gives wrong answers
• They worry about patient privacy and shaky data protection
• They see a real chance of clinical mistakes if tools are used without training

Blease warned that, "The real risk isn't that GPs are using AI. It's that they're doing it without training or oversight." Her point hangs heavy. With no clear guardrails, even small errors can snowball into big problems.

The study offered another twist. Policy makers hope AI will free up doctors so they can see more patients. Yet many GPs say the extra time mostly goes to rest and recovery. They use it to reduce overtime and keep burnout at bay. It is a quiet reminder that technology can ease pressure, but it cannot solve staff shortages on its own.

The context

The rush toward AI follows a tough spell in British primary care. Long waits for GP appointments push both doctors and patients to try new options. Patients themselves are experimenting with AI. Healthwatch England found that nine percent use AI tools for health information, often when they cannot get a GP appointment. But the quality of advice swings wildly. One patient even received an answer that mixed up shingles with Lyme disease.

This growing public use only sharpens the need for proper rules. Both studies stressed the lack of national oversight. Without it, GPs are left to guess which tools are safe. Fisher called it a "wild west" for a reason. The government has set up a new commission to fix this and aims to shape a system where AI is used safely, openly, and with proper accountability.

AI is already in the room. The question now is whether the system can catch up before the tech races further ahead.