The National Commission into the Regulation of AI in Healthcare report aims to strengthen the regulation of AI in healthcare, so it remains safe, keeps pace with innovation and continues to earn people's trust. 

Professor Henrietta Hughes, NHS GP, deputy chair of the National Commission and Patient Safety Commissioner for England, said: ‘Patients have told us that to trust AI is safe, they want to know when it's used, that it supports rather than replaces clinicians, and that there is clear accountability if things go wrong. That's exactly what our recommendations deliver.'

Further recommendations include: continuous, real-world monitoring throughout an AI model's working life; easy public access to information on AI-enabled medical devices; and stronger powers for the MHRA, including enhanced enforcement.

The independent commission was established by the MHRA in September 2025 and includes diverse expertise across health, regulation and AI, including international contributors from around the world, including the US and Singapore.  

Reaction

Dr Stephen Harden, president of the RCR, said: ‘This commission has come at exactly the right moment. The AI landscape is only getting more complex, and so we are proud to have contributed to this important work. We are particularly pleased to see specific support for monitoring of AI-enabled medical devices once they are in use; getting this right will give doctors confidence to use AI tools, while protecting patient safety.'

Dr Jennifer Dixon, chief executive of The Health Foundation, said: ‘The commission's recommendations are an important step towards the goal of making the NHS a world-leading AI-enabled care system. But to realise that ambition and deliver benefits to the NHS and patients, policymakers now need to strengthen the infrastructure to harness AI's potential.'

The Health Foundation has today published research finding the UK public is open to the use of AI in health care, but want strong safeguards: tools must be accurate, subject to meaningful human oversight, proportionately regulated according to risk, and must not lead to worse care for any group.

Dr Sarah Townley, deputy medical director at Medical Protection, said: ‘We particularly welcome the Commission's recognition of the risk of clinicians becoming a ‘liability sink' when AI goes wrong and agree with their observation that healthcare professionals should not be held accountable for risks they cannot realistically manage. Our Closing the AI Liability Gap report argues that responsibility should sit with those best able to control or mitigate the risks.

‘However, it's disappointing to see that the report fails to make specific recommendations on updating existing product liability laws. Clearer, fairer liability would protect doctors from potentially taking the blame for software errors, encourage developers to prioritise safer design and increase confidence in using AI tools, unlocking its potential across the healthcare sector.'

Dr Satya Raghuvanshi, vice-president of clinical at Accurx and former NHS doctor, said: ‘Accurx Scribe is subject to ongoing clinical audit, a random sample of notes are reviewed monthly for accuracy and safety, and every issue is fed straight back into the product. Clinicians still review and sign off every note before it reaches the record. That's the standard responsible AI in healthcare has to meet, and it's what the commission is rightly calling for. We look forward to the Government's response.'