Info Gov

Several police forces in England and Wales have been told to stop using artificial intelligence tools to prepare court statements and carry out other criminal justice tasks, after concerns that inaccurate outputs could undermine legal proceedings.

Alex Murray, head of the government's newly established Police.AI centre, told the Financial Times (FT) that he had intervened at a number of forces that had begun deploying commercially available AI systems before they had been properly assessed. In an interview published by the FT on 6th June 2026, Murray said that he had instructed some forces to pause their use of the technology, saying safeguards had to be in place before widespread adoption could proceed.

Any AI technology used in the criminal justice system needed to meet a standard of accuracy "beyond reasonable doubt", Murray said. He said that Police.AI had stepped in after some forces began using AI to help officers convert interview recordings into court-ready statements. He also warned forces against deploying the technology in other sensitive areas, including the preparation of disclosure schedules (which set out evidence that must be shared with the defence before trial), until appropriate checks had been completed.

"We've said to some police forces, 'You can't do that, because we haven't gone through all the checks and balances," he said.

Murray acknowledged that the risks involved were manifested when West Midlands Police previously relied on AI-generated material produced by Microsoft's Copilot that fabricated details about a past match involving Maccabi Tel Aviv, as part of a dossier used to support a proposed ban on the club's supporters attending a fixture against Aston Villa. The incident drew widespread criticism and highlighted the danger of AI “hallucinations”, where systems generate false but plausible-sounding information.

Murray said all forces now have policies requiring staff to check everything produced by Copilot.

Despite the concerns, Murray argued that properly assessed AI systems could deliver significant benefits for forces dealing with growing workloads and increasingly complex investigations.

Police.AI was established in April this year as part of government efficiency reforms and provided with £115 million of funding over three years, with the aim of accelerating the responsible use of AI across all 43 forces in England and Wales. According to Home Secretary Shabana Mahmood, it is intended to generate time savings equivalent to adding 3,000 officers to the roughly 145,000 currently serving in England and Wales.

Among the applications Murray highlighted was the use of AI to rapidly scan CCTV footage for suspects, and a tool being developed to identify and classify images found on seized devices in child sexual abuse investigations, reducing the volume of harmful material that officers would otherwise need to review themselves.

"I think the benefit that automation offers, with the appropriate guardrails, policy and training, outweighs the disadvantages," Murray said.

Also in this section

Jun 11, 2026

AI liability gap leaves NHS and clinicians exposed to negligence claims, MPS says

The Medical Protection Society (MPS) has called on the government to legislate to classify AI systems as products, warning that a growing mismatch between healthcare AI and the UK's product liability framework risks leaving the NHS and individual clinicians to absorb the full legal cost when defective systems cause patient harm.
Jun 01, 2026

ICO sets out plans for a statutory AI code of practice and procurement guidance

The Information Commissioner’s Office (ICO) has set out plans to develop a statutory code of practice on artificial intelligence and automated decision-making, alongside new guidance to help organisations – including public bodies – ensure good data protection practice when procuring AI tools.

InfoGov Masthead Newsletter 800