The MHRA has launched a new regulatory sandbox to test how artificial intelligence can be used to make medicines safer and accelerate their development.

Announced by Science Minister Lord Vallance during London Tech Week on 9 June 2026, the Regulatory Innovation Office-backed initiative will explore how AI can improve the assessment of accuracy and safety, better predict risks, and detect effects that existing approaches may not capture.

The regulator said the programme responds to a significant problem in medicines development – adverse drug reactions cost the NHS more than £2bn annually and send about 250,000 people to hospital each year, while about 90% of drugs fail during development because existing methods cannot reliably predict how medicines will behave in people. 

Up to five AI-driven approaches will be tested from summer 2026, with the sandbox also exploring how better use of clinical data can improve understanding of how medicines work across diverse patient groups. This supports the UK Government's AI for Science Mission and its ambition for Britain to have the world's most AI-enabled healthcare system.

Gerard Hanratty, Partner and Head of Health and Life Sciences at Browne Jacobson, said: "We have long argued that establishing clear ethical guidelines, and rigorous testing and evaluation, must be at the heart of any AI regulatory framework. The regulatory sandbox model – giving companies a controlled environment in which to test AI tools alongside regulators – is something we've seen work well in other jurisdictions, most notably the UAE, and it's encouraging to see the MHRA now adopting it in a medicines context.

“The key to successfully regulating AI lies in striking the right balance between promoting innovation and ensuring accountability, as well as building a genuine evidence base before mandating or restricting specific uses. 

“Legal frameworks are at their most powerful when used as enablers rather than barriers, and this sandbox creates a clear pathway for the MHRA's modernisation agenda. It should give companies the confidence to invest and innovate in the UK.

“What is particularly welcome is the commitment to explore how better use of clinical data can improve our understanding of how medicines work across different patient groups, including children, older people and those from diverse backgrounds who are so often under-represented in clinical studies. 

"Diversity and representativeness in patient populations is a fundamental issue in clinical trials because it affects the reliability of safety data and, ultimately, the equity of medicines development. Addressing it through AI-driven tools could be transformative.”