A UK commission wants medical AI regulated as it changes, not approved once and left alone
After a year-long review and evidence from more than 12,000 people, a commission chaired by NHS doctors proposes replacing one-time device approval with staged authorization and continuous real-world monitoring.
The UK's National Commission into the Regulation of AI in Healthcare, chaired by Prof. Alastair Denniston and established by the MHRA, published its blueprint on September 10, 2026, after a year-long review that gathered evidence from more than 12,000 people. Its central recommendation replaces one-time device approval with a staged authorization model (described as an "L-plates" system, referencing the UK's provisional-driver-license scheme) paired with continuous real-world monitoring and stronger MHRA enforcement powers.
Why one-time approval doesn't fit this category of product
Medical devices have historically been approved once, against a fixed specification, and left alone. That model assumes the product doesn't change after approval -- which doesn't hold for an LLM-based tool whose behavior can be updated, fine-tuned, or shifted based on the data it sees in deployment. Prof. Denniston framed the goal as building "a future healthcare system that is increasingly tech-enabled and always people-centred," which only works if the regulatory model tracks a system that keeps changing after it ships.
A template that reaches past healthcare
The mechanism, staged authorization plus continuous monitoring instead of a single approval gate, is a concrete answer to a problem every AI regulator eventually has to solve, including those outside medical devices. For enterprise buyers in any industry, it's an early look at what "continuous monitoring" as a formal regulatory requirement, rather than an internal best practice, might eventually look like in a vendor contract.