Introduction
The UK Government is undertaking significant reforms in the regulation of medical devices, with a particular focus on software and AI products. This initiative is driven by the need to ensure patient safety, promote innovation, and respond to the recommendations from the Regulatory Horizons Council (RHC).
Description
The UK Government is reforming the regulation of medical devices, with a particular emphasis on software and AI products, in response to the 15 recommendations from the Regulatory Horizons Council (RHC). This initiative aims to facilitate the safe and swift development of AI medical devices while ensuring patient safety and promoting innovation. The Government’s position has been articulated by the Parliamentary Under-Secretary of State for Patient Safety, Women’s Health and Mental Health in a letter.
To support manufacturers in compliance [2], the Medicines and Healthcare products Regulatory Agency (MHRA) is enhancing its regulatory capacity with long-term funding commitments. This includes a £1 million investment for the AI Airlock [3], the first regulatory sandbox for AI as a Medical Device (AIaMD) [2], which will operate until spring 2025. The AI Airlock aims to identify regulatory challenges and assist manufacturers in gathering evidence for product approval [4]. Additionally, the MHRA is developing regulatory guidance that emphasizes safety, innovation [2] [3] [4], and positive patient outcomes [2], while also addressing the unique challenges posed by AI technologies [3].
A pilot Scientific Dialogue Programme has been initiated to help innovators refine their evidence-generation strategies [4], as robust clinical validation data is essential for the regulatory assessment of AI/ML-based digital health solutions [4]. The MHRA emphasizes the need for accuracy [4], reliability [4], and clinical relevance in AI/ML algorithms [4]. A review of generative AI in healthcare has been commissioned to alleviate pressures within the NHS and promote the adoption of AI through innovative regulatory measures. Manufacturers are required to evaluate and mitigate risks related to AI generalisability and bias [2], providing documented evidence for their products [2]. The Software and AI as a Medical Device Change Programme is dedicated to addressing safety and effectiveness concerns [2], ensuring trust in AIaMD among users and patients [2].
The regulatory framework will support mechanisms for accelerated access and post-market evidence generation [2], with legislative reforms enhancing post-market surveillance to improve patient safety [2]. Manufacturers are expected to collaborate with health institutions to demonstrate safe performance in local settings prior to deployment [2] [3]. Additionally, stakeholders are encouraged to work together to establish standards for post-market monitoring of performance and safety [3], including contractual agreements on monitoring approaches [3].
Predetermined change control plans (PCCPs) will allow pre-approved modifications to software without further regulatory review [2]. The MHRA is investing in accessible guidance to facilitate safer access to medical devices and is building relationships with international regulatory bodies to streamline access and reduce regulatory duplication. An international reliance framework is being developed to ensure safe access to medical devices.
As part of the AI Airlock initiative [3], the MHRA has selected five new technologies to better understand the regulation of AI-powered medical devices [3], focusing on critical areas such as cancer and chronic respiratory disease [3]. A public-facing register will enhance transparency regarding AIaMD on the market [2], including risk classifications and intended uses [2].
Greater patient and public involvement in the design and regulation of AIaMD is advocated [2], as the UK aims to lead internationally in this area. The Government emphasizes the need for a regulatory framework that is light on legislation [3], leveraging existing regulations for software as a medical device while addressing the unique challenges posed by AI technologies [3]. Furthermore, the UK Government’s investment in genomic medicine may raise additional regulatory and data concerns as genomic data is utilized on a larger scale [4], indicating an anticipated evolution in the landscape of digital health technologies alongside ongoing adjustments to regulatory frameworks.
Conclusion
The reforms in the regulation of medical devices [1] [3], particularly those involving AI and software, are poised to significantly impact the healthcare landscape in the UK. By prioritizing patient safety and innovation, the UK Government aims to lead internationally in the regulation of AI medical devices. These efforts are expected to enhance the development and deployment of AI technologies in healthcare, ensuring they are safe, effective [2], and beneficial for patients.
References
[1] https://www.gov.uk/government/publications/the-regulation-of-artificial-intelligence-as-a-medical-device-government-response-to-the-rhc
[2] https://www.gov.uk/government/publications/the-regulation-of-artificial-intelligence-as-a-medical-device-government-response-to-the-rhc/the-regulation-of-artificial-intelligence-as-a-medical-device-government-response-to-the-regulatory-horizons-council
[3] https://htn.co.uk/2025/03/11/government-accepts-15-recommendations-on-the-regulation-of-ai-as-a-medical-device/
[4] https://iclg.com/practice-areas/digital-health-laws-and-regulations/united-kingdom
