Our Impact

Recognising the urgent need for leaders of AI implementation across the healthcare workforce, we created the AI Implementation (Healthcare) MSc, the UK’s first taught programme focused on equipping clinical, technical and operational leaders with the expertise to advance responsible and impactful AI innovations. In parallel, online seminars and CPD courses have been developed to accelerate shared understanding across clinicians, digital

Delivered by Hardian Health, this project has established a tool that standardises the availability of information about medical devices. HaRi displays registration data from the UK, EU and USA as well as peer reviewed evidence and adverse event data from the UK and UK. Designed for regulators, providers, innovators, procurement teams, and patients, it enables faster, safer adoption through its free-to-use site for public users and additional functionality for commercial users. We’re exploring how further enhanced functionality for the MHRA could support global intelligence including advanced early signal detection of safety.

Funded by the Health Foundation, this practical self-assessment tool will help UK healthcare organisations prepare for safe, effective, and equitable AI adoption. It evaluates readiness across key domains including data and digital infrastructure, governance, workforce capability, ethics, and organisational strategy. The team in Birmingham’s Responsible Innovation Group for Health and Technology (Bright) developed this tool through desk research, expert interviews, and stakeholder consultation. The checklist is now open for public consultation and is being piloted with NHS trusts.

The Borderline Manual for SaMD and AIaMD provides practical UK guidance on classifying AI health technologies as medical devices. Developed by the University of Birmingham and Hardian Health, it clarifies regulatory status, risk classification and compliance pathways to support safe, accelerated innovation. We’ve worked closely with the regulator, MHRA, to support consistent borderline classification decisions. Our aim is to reduce regulatory ambiguity for innovators and accelerate safe adoption.

Funded by NHS England’s Data for R&D Programme and the Office for Life Science, this project explored how SDEs can be made secure, innovation-ready, and trusted. Covering governance, data access, evaluation, disclosure risk, and sustainable operating models for pre- and post-market AI testing, we published an analysis of model egress and disclosure risk, recommendations, a report and insights from patients and the public. Our vision is a future-ready SDE ecosystem that balances security with functionality, enabling access to NHS data for AI innovation, robust governance, and transparent engagement with the public.

An international team of academics, health professionals and technologists, led by Dr Joe Alderman, began to build a global effort on safety guidelines for the public when using AI health chatbots by building The Health Chatbot Users’ Guide in February 2026

This project evaluates a generative AI tool that produces radiology-style reports in three imaging modalities. Led by the University of Glasgow, the project assesses clinical accuracy, safety and acceptability to support safe NHS Adoption and regulatory guidance.

We delivered a workshop at the Health AI Global Governance Forum in Nairobi in December 2025. The session explored our framework for responding when Health AI fails, based on the OODA Loop (Observe, Orientate, Decide, Act), supporting rapid, structured decision-making for clinical AI solutions.

Alastair Denniston, Director of CERSI-AI, was on the Davos 2026 programme for the Roundtable: From Pilots to Impact – Collaborative Pathways for Responsible AI in Health. Alastair spoke of the importance of the “value intersect”: the point where priority clinical need, technological capability, and system readiness align. Strengthening regulatory clarity and preparedness is a critical part of that readiness, and a central focus of the National Commission’s work.