Prime Highlights
- The MHRA is launching an AI regulatory sandbox to help identify drug safety risks earlier and accelerate the development of new treatments.
- The programme will allow companies, researchers and regulators to test AI tools designed to improve medicine safety assessments and decision-making.
Key Facts
- Around 90% of drug candidates fail during development, highlighting the need for better predictive tools.
- Adverse drug reactions lead to approximately 250,000 hospital admissions annually in the UK and cost the NHS more than £2 billion a year.
Background
Britain’s medicines regulator, the Medicines and Healthcare products Regulatory Agency, is setting up an AI regulatory sandbox to test tools that could spot drug safety risks earlier and speed up the development of new treatments.
The programme was unveiled by Science Minister Lord Vallance on 9 June 2026. Up to five AI technologies will go through testing in the first phase, with work scheduled to begin in summer 2026. The MHRA will run the programme alongside industry and academic partners.
The sandbox gives companies and researchers a controlled space to trial AI tools built to predict how medicines behave in people and flag potential risks that standard testing methods tend to miss.
The programme sits separately from the MHRA’s existing AI Airlock scheme, which focuses on AI medical devices rather than medicines development. Lawrence Tallon, Chief Executive at the MHRA, said the timing reflects a broader shift in what science can now offer. According to him, these technologies have the potential to enhance drug safety research, provide better evidence for decision-making and speed up the development of new therapies.
The case for change is backed by hard numbers. Around 90% of drug candidates fail during development, largely because current methods cannot reliably predict how treatments will perform in patients. At the same time, adverse drug reactions send roughly 250,000 people to hospital in the UK each year and cost the National Health Service more than two billion pounds annually.
Preet Gill said the programme will allow innovators and regulators to work together in testing new technologies, helping generate the evidence required to deliver safer treatments more quickly. She added that the process should also reduce reliance on animal testing and make medicines development more efficient.
The sandbox is part of the government’s wider AI for Science programme and draws funding from the Regulatory Innovation Office. It will explore advanced modelling, synthetic data, and alternatives to animal testing as the UK pushes to modernise how medicines are assessed and approved.
