The UK government has accepted all 44 recommendations from the National Commission into the Regulation of AI in Healthcare and says an implementation plan and roadmap will be published by Spring 2027. The response sets priorities for changing how AI-enabled medical devices are assessed, monitored and used in healthcare.

The commission said current regulatory approaches were largely designed for products that remain static and can be assessed reliably at one point. AI-enabled products may change rapidly and perform differently across settings, depending on data, workflows, people and organizations.

James Frith, parliamentary under-secretary of state for health innovation, said the government agrees the UK needs a proportionate, risk and lifecycle-based approach. He said oversight should take a system-wide view and demonstrate benefits in real-world settings to support adoption and trust.

Over the next year, the government will prioritize modernization of the rules for AI-enabled medical devices. That work includes clearer guidance on when a product qualifies as a medical device, an improved classification approach, and guidance on how intended purpose should account for design and functionality.

Monitoring and Responsibility Across the Lifecycle

The response also commits to promoting equity throughout the lifecycle of AI-enabled technologies. It calls for stronger post-market assurance, new surveillance mechanisms and improved monitoring of safety and performance changes in real-world settings.

The government will explore staged authorization pathways that could allow earlier access while generating real-world evidence. It also plans to expand regulatory sandboxes, where new technologies and regulatory approaches can be tested.

The MHRA will develop guidance on how manufacturers should identify and communicate the operating conditions needed for safe deployment and use. The guidance is expected to include practical information and illustrative examples of risk controls.

The DHSC and NHSE will work on an AI readiness toolbox to help healthcare organizations prepare for AI-enabled technologies. A working group comprising the DHSC, MHRA, CQC and equivalent organizations will focus on reporting, monitoring and learning from real-world use.

For trust and transparency, the MHRA and its partners plan to increase patient and public engagement in regulatory decisions. They also intend to offer developers and manufacturers clearer guidance and opportunities for early engagement, while exploring ways to communicate information about AI technologies and safety concerns.

The DHSC will work with healthcare partners to improve transparency and consistency in telling patients when AI-enabled technologies are used in their care. The government also plans a public-facing database or tool where people can search for information on adverse incidents involving medical devices.

Further Rules and Delivery Dates

A cross-system programme board with representatives from key partner organizations will oversee delivery. It will meet quarterly to review progress against objectives and assess emerging risks.

The implementation plan and roadmap, due by Spring 2027, will add detail about roles, responsibilities and key deliverables. The MHRA also plans to launch a consultation by Spring 2027 on a new approach to classifying and qualifying software and AI-enabled devices.

After that consultation, the government intends to use secondary legislation to introduce an updated definition, a classification system that better reflects a proportionate approach to device benefits and risks, and an updated definition of intended purpose.

Other accepted recommendations include guidance on cybersecurity expectations for software and AI-enabled devices and exploring ways to automate low-burden reporting by embedding it in systems such as EPRs. The government also promises a coordinated effort to improve AI literacy across the workforce, combining a shared baseline with learning tailored to roles and responsibilities.

The decision covers the commission’s recommendations; the roadmap and consultation remain future steps. Readers can compare this healthcare process with U.S. crypto rulemaking, regulation and tokenization, and crypto’s financial use for other coverage of regulation and technology.

Read the government response details for the source account of the accepted recommendations.