The United Kingdom is preparing a new approach to the use of artificial intelligence in healthcare after an independent commission published recommendations on 10 September 2026 on how AI systems could be authorised, supervised and monitored across the NHS and the wider healthcare sector. The document does not introduce new law: it is advice for the regulator, the government and organisations in the health system.

The National Commission into the Regulation of AI in Healthcare was established by the Medicines and Healthcare products Regulatory Agency (MHRA) in September 2025 as an independent, non-statutory advisory body. It examined how regulation should adapt to technology that can evolve faster than conventional medical devices. Its members include specialists in medicine, regulation and AI, with international contributors including from the United States and Singapore.

The Commission argues that AI could bring substantial benefits for patients and staff, but that AI models cannot always be regulated in the same way as traditional devices. Some systems can change, adapt or perform differently depending on the clinical context. One of its central recommendations is therefore staged authorisation, which the MHRA has compared to learner-driver “L-plates”.

A new AI system could first be deployed in a tightly controlled setting, with closer supervision and clear limits on use. If it then demonstrates safe and effective performance in real-world conditions, restrictions could be eased and authorisation broadened. The stated aim is faster access to promising technologies without removing safeguards. That does not mean every AI system already used in the NHS would immediately operate under this model: it is a proposal for a future framework.

A second major recommendation is a shift from a single, point-in-time approval towards continuous monitoring throughout a system’s working life. Models can evolve after deployment, and their behaviour may vary with patient populations, data or clinical settings. Regulators should therefore keep receiving information about performance after the technology enters hospitals and other services.

Human oversight is a central theme. People consulted during the review indicated that public acceptance of AI in healthcare depends on the technology supporting clinicians rather than replacing them. Patients also want to know when AI is used in diagnosis, treatment or another healthcare decision. The Commission says accountability must be clear if an AI-generated recommendation contributes to an incorrect decision — whether that involves the developer, the healthcare organisation, the professional using the system, or a combination of those parties.

The recommendations rest on a wide engagement programme. According to the MHRA, evidence and views were gathered from more than 12,000 people over a year, including patients, members of the public, clinicians, healthcare leaders, industry and technology developers. The agency describes it as the largest engagement of its kind yet undertaken in the UK on the regulation of healthcare technology.

Additional research by the Health Foundation, with Ipsos, examined public attitudes in more depth. Principles that emerged include accuracy as a leading priority, human oversight as a critical condition, regulation proportionate to risk, and the requirement that AI should not result in poorer care for particular groups of patients.

AI is already in use in the NHS. The MHRA points to examples including systems that help identify serious conditions such as strokes and skin cancers earlier, and voice-enabled tools that reduce time spent on documentation. The Commission argues that wider adoption should still be accompanied by clearer rules on responsibility, transparency and ongoing performance monitoring.

The UK government and the MHRA will now consider the recommendations. A separate cross-government response is expected in due course. Until then, the 10 September report is a blueprint — not a set of rules already in force — for a technology that is becoming a larger part of modern healthcare.

Photo: National Cancer Institute / Daniel Sone. A healthcare professional using a laptop, with a stethoscope on the desk. The image is not from the NHS and is not from the Commission.

Source consulted: Independent Commission led by NHS doctors sets out blueprint to accelerate safe AI adoption in healthcare | GOV.UK / MHRA; Recommendations for a future regulatory framework | National Commission into the Regulation of AI in Healthcare; The public’s views on the regulation of AI in health care | Health Foundation.