New @probablefutures.bsky.social report on #AI in law enforcement & criminal justice. Use of AI is running ahead of safeguards - putting a 'Human in the loop' can give false assurance figshare.northumbria.ac.uk/articles/rep... and our mapping tool is at probablefutures.github.io/ai-mapping/m...
figshare.northumbria.ac.uk
The Use of Probabilistic AI in Law Enforcement and the Criminal Justice System of England and Wales
This report provides a detailed account of how probabilistic AI is being introduced, understood, and operationalised across law enforcement and the wider criminal justice system of England and Wales. Drawing on semi-structured interviews with professionals from policing, prosecution, the Bar, government, oversight bodies, and the technology sector, it examines perceptions, their practical use, and the gap between intended design and user experience. The report examines implementation challenges, including the risks arising when AI outputs feed into further automated processes, while also assessing the operational benefits AI can deliver when deployed responsibly. It further considers the funding and innovation structures shaping adoption, evaluates the adequacy of current institutional, evidential, and governance arrangements, and concludes with recommendations for the bodies responsible for overseeing these technologies.The central finding is twofold. AI systems are already generating significant gains in efficiency, decision-making, and, in some contexts, public protection, particularly where deployment has been problem-led, iterative, and properly evaluated. At the same time, adoption is advancing more rapidly than the institutional, evidential, and governance frameworks needed to support it. The result is an increasing gap between technological capability and the conditions required for its legitimate, reliable, and responsible use.