AI in Medicine: How a Neural Network Triages Patients in the NHS App

In July 2026, the UK government built a neural network into the NHS App. The system will assess the symptoms of patients in England and determine whether they need to see a general practitioner, visit a pharmacy or call an ambulance.

How AI triage works in the NHS App

The new tool is built directly into the NHS App, the official mobile app of the National Health Service. The user describes their symptoms, and the algorithm analyzes them and routes the patient down one of three paths: a general practitioner (GP), a pharmacy or the emergency department (A&E).

The mechanics are simple but significant. For the first time, the initial routing decision is made not by a human operator but by an AI embedded in one of the country's most widely used health apps. In effect, the system performs the sorting function that used to fall to phone operators at the 111 and 999 services. Automating this step is meant to relieve pressure on both call centers and GPs themselves, who now often see patients who could simply have bought a sore-throat remedy at a pharmacy.

Key parameters of the pilot:

• Coverage: 200,000 patients in England during the first year
• The AI is built into the official NHS app
• Three routes: GP, pharmacy, ambulance / A&E
• The initiative is part of the £10 billion NHS modernization program

Why the NHS is investing £10 billion in digitization

For decades, the NHS has operated under chronic systemic overload. Waits for GP appointments stretch to weeks, ambulance services are swamped by inappropriate calls, and the 111 phone lines, the first point of contact for millions of patients, are constantly overloaded. By the NHS's own estimates, a significant share of contacts did not require medical care in the first place: the person needed a pharmacy or simply information about their symptoms.

The £10 billion technology plan is one of the largest government IT investments in healthcare in Europe. It aims to make digitization a systemic answer to these pressures rather than a collection of scattered experiments. Besides AI triage, the program includes a move to digital medical records, data integration between hospitals and primary care, and replacement of outdated IT infrastructure.

The NHS App is already a mature platform with tens of millions of users. Through it, people in the UK book appointments, get test results, order repeat prescriptions and check their vaccination status. Adding an AI layer is meant to turn the app from a booking tool into a full first point of help for patients: a system that does not merely log a request but also makes the initial clinical decision on routing.

What this means

Primary care is one of the most overloaded segments of healthcare in any country, and this is where AI can deliver the greatest practical benefit. If the 200,000-patient pilot shows that AI triage really does reduce the share of inappropriate contacts, cut the load on GPs and speed up access to the right care, the NHS will have grounds to scale it across all of England, which means tens of millions of app users.

The British experience of deploying AI at the state level in primary care will be closely studied outside the UK: there are still only a handful of major examples of a national health system systematically handing patient routing over to AI. If the NHS succeeds, it will become a benchmark for similar reforms worldwide.

How does the neural network decide whether you need a doctor?

The system assesses the patient's symptoms and determines whether they need a GP visit, pharmacy help or an ambulance.

Can a neural network replace a doctor in the NHS?

The neural network does not replace the doctor. It helps direct the patient to the right specialist or service. It is a prioritization system (AI triage).