NHS App Gets AI Triage Tool. Here Is What It Actually Does and What the Critics Want Answered.

What the Tool Does

NHS England announced that an AI-powered triage feature is coming to the NHS app for patients in England. The system asks a series of questions and then directs users to a GP appointment, pharmacy, A&E, a community service, or self-care guidance — whichever fits best.

The rollout target, according to NHS England, is more than 200,000 patients over the next 12 months, with full availability to all app users by April 2028.

The Pilot Numbers

A trial at Wealden Ridge Medical Partnership in Sussex produced a 29% reduction in patients queuing on the phone for an appointment.

Dr. Ragu Rajan, a GP at the practice, described the tool plainly: "It hasn't replaced our judgement — it's given us back the time to use it."

Sir Jim Mackey, chief executive of NHS England, said the goal is to "help get patients to the best service for their needs first time... so that clinicians can make sure those most in need of a GP appointment can get one sooner."

Part of a £10 Billion Tech Overhaul

The AI triage tool is one piece of a broader £10 billion investment the UK government allocated in 2025 to overhaul the NHS's technology, digital, and data systems.

Also included in this wave: an England-wide rollout of AI notetaking tools that record conversations between patients and NHS staff, then generate real-time transcriptions and clinical summaries. The initial phase targets hospital outpatient appointments at four NHS trusts in and around London — St George's, Epsom and St Helier, Croydon, and Kingston and Richmond.

Two additional trusts, Alder Hey Children's NHS Foundation Trust in Liverpool and Manchester University NHS Foundation Trust, are expanding AI notetaking programs they already run.

A trial led by Great Ormond Street Hospital for Children, conducted across nine NHS sites in London, found that staff spent almost 25% more time interacting with patients when using the notetaking technology.

The Legitimate Concerns

The strongest objection to this rollout is NOT that AI is inherently dangerous in healthcare. It is that automated triage systems can fail silently, and the patients most likely to be misrouted are often the ones least equipped to push back. Elderly patients, people with complex comorbidities, those with limited digital literacy, and non-native English speakers could all be steered toward the wrong care pathway if the tool's question set is not robust or its outputs are not properly validated across diverse populations.

Prof. Lynn Woolsey, chief nursing officer at the Royal College of Nursing, called the rollout "an important step in upgrading technology in the NHS" but stated plainly that patient safety and confidentiality must be at the "heart of any AI triage system, with a guarantee that a health professional will be the one making decisions at key points in that process."

Pritesh Mistry, a fellow at the King's Fund think-tank, said the announcement "could help turbo-charge improvements in how [the] NHS uses modern technology," while still urging that inclusion and access remain a priority as the system scales.

What This Actually Requires to Work

The 29% reduction in phone queue demand at Wealden Ridge is promising, but one practice in Sussex is NOT a large-scale validation. The April 2028 full-deployment deadline gives NHS England roughly 21 months from now to scale from a few hundred thousand users to the entire app user base — a population in the tens of millions.

Key questions that remain open: Will the AI triage tool be audited for accuracy by patient group and region? Who is liable when a patient is triaged to self-care and deteriorates? How will NHS England handle patients who cannot or will not use the app?

NHS England has NOT published a public protocol on clinician override requirements or on how error rates will be tracked at scale. Until that exists, the RCN's demand for a guaranteed human decision point at critical junctures should be a formal operational requirement, not a general aspiration.