AI-assisted tool on FDP will help write hospital discharge letters

The NHS federated data platform will host a new AI-assisted discharge summaries tool, the government has announced. 

The technology, being developed at Chelsea and Westminster NHS Trust, uses a large language model to extract key details from medical records, such as diagnoses and test results, to help write the documents needed to discharge people from hospital.

After being reviewed by a medical expert, the documents would be used to discharge a patient from a ward and refer them to other care services that may be needed.  

Wes Streeting, health secretary, said: “This potentially transformational discharge tool is a prime example of how we’re shifting from analogue to digital as part of our 10 year health plan.

“We’re using cutting-edge technology to build an NHS fit for the future and tackle the hospital backlogs that have left too many people waiting too long.

“Doctors will spend less time on paperwork and more time with patients, getting people home to their families faster and freeing up beds for those who need them most.”

A press release, published by the the Department of Science, Innovation, and Technology on 17 August 2025, said that the technology could cut waiting lists by enabling information to be handed over to different care services more efficiently and allowing patients to get home more quickly.

Responding to the initiative, Matthew Taylor, chief executive at the NHS Confederation, said: “We welcome this announcement which shows how AI can boost efficiency and help NHS staff discharge patients quickly.

“When rolled out effectively, AI has great potential to improve NHS productivity by helping to reduce waiting lists.

“It is positive that through this pilot project, the AI-assisted tool will reduce the administrative burden on doctors and capture critical information by helping them draft discharge documents faster.

“But for AI tools to work well, sufficient time is needed to obtain consent and verify accuracy as the health worker remains responsible for both the precision of the information and patient confidentiality. 

“This will ensure high quality documents are produced and patient trust is maintained.”

The technology is one of several projects which will be developed and trialled as part of the AI Exemplars programme, with those showing the most promise potentially rolled out more widely.

Other projects include the suite of AI tools known as ‘Humphrey’, intended to make the civil service more efficient, such as ‘Consult’ which analyses the thousands of responses that government consultations receive in hours and creates interactive dashboards so that policymakers can explore what the public are saying.

Speaking on a visit to Chelsea and Westminster Hospital, Peter Kyle, technology secretary, said: “This is exactly the kind of change we need: AI being used to give doctors, probation officers and other key workers more time to focus on delivering better outcomes and speeding up vital services.”

The plans deliver on the AI Opportunities Action Plan, announced in January 2025, which recommends rapidly testing, building or buying tools focused on public sector productivity.