mhra

UK accepts 44 healthcare AI recommendations as AI Airlock Phase 3 opens

The government has accepted the Commission’s healthcare AI recommendations in full. MHRA’s next AI Airlock phase will focus on post-market surveillance and lifecycle regulation.

Ersin ÇETİNMDR Auditor
Türkçe oku
Conceptual illustration of lifecycle monitoring and regulatory oversight for AI-enabled medical devices
AI-generated conceptual illustration, not a photograph of an actual event.

What Changed

The UK government announced on 6 October 2026 that it accepted all 44 recommendations of the National Commission into the Regulation of AI in Healthcare. The proposed direction emphasises assessment and monitoring throughout an AI-enabled medical device’s working life, rather than over-reliance on a one-time assessment.

Alongside the response, the MHRA opened applications for Phase 3 of AI Airlock, its regulatory sandbox. This phase will focus on post-market surveillance and lifecycle regulation. The programme has secured three years of additional government funding.

Who Is Affected

The announcement concerns developers and manufacturers of AI-enabled medical devices, healthcare organisations deploying them, and patients. Eligible developers can apply to a programme bringing innovators, regulators and healthcare partners together to investigate regulatory challenges using real products.

Wider commitments include strengthening patient participation in decisions, keeping patients informed when AI is used in their care, and improving access to redress where care standards fall below expectations.

Regulatory and Technical Implications

The MHRA plans draft guidance on managing changes to AI-enabled devices as they adapt and improve. Qualification and classification will be addressed through a separate consultation.

The response also commits to exploring staged authorisation pathways, potentially allowing promising tools into the NHS earlier under close supervision while further real-world evidence is collected. These are policy and development commitments: accepting the recommendations should not be equated with new device requirements already being in force.

Although the release describes UK-wide delivery, the Scottish Government says it will set out Scotland’s response in due course. Identical implementation across all administrations should therefore not be assumed.

Manufacturer Actions

Developers considering participation can consult the official programme information for eligibility and application arrangements and attend the applicant webinar. The supplied release does not state an application closing date.

Manufacturers can monitor the forthcoming change-management guidance, qualification and classification consultation, and implementation roadmap. Evidence of performance and safety throughout deployment is central to the announced direction.

Confirmed Dates

The Commission published its recommendations on 10 September 2026. Government acceptance and Phase 3 applications were announced on 6 October 2026. The webinar is scheduled for 22 October 2026 at 10:00, with no time zone specified in the supplied text. Selection of the first participants is planned for November 2026.

Draft change-management guidance is promised by December 2026. The qualification and classification consultation is planned for early 2027, with the full implementation roadmap due by spring 2027.

Official Source

GOV.UK: Government backs recommendations of NHS doctors-led AI Commission. Published and updated on 6 October 2026. A same-day correction changed the number of priority areas from nine to ten.

https://www.gov.uk/government/news/government-backs-recommendations-of-nhs-doctors-led-ai-commission

Source and related services

Government backs recommendations of NHS doctors-led AI Commission ↗

Source date: 06/10/2026

This news provides general information and does not replace device-specific legal or conformity assessment advice.

Related news

MHRA opens AI Airlock Phase 3 applications with focus on post-market surveillance →MHRA updates contact address on good machine learning practice guidance →MHRA updates medical-device exceptional use authorisation lists →
Contact an MDR Expert