HM Courts & Tribunals Service
Unlocking public sector productivity with AI innovation
Tags
After a challenging few years, the UK legal system is facing significant pressure to ensure swift access to justice. Courts and tribunals across England and Wales, including non-devolved tribunals in Scotland and Northern Ireland have finite resources, high demand, and an often-complex case mix. To address this, they must build on reformed digital services and newly stabilised digital platforms to continuously find ways of enhancing productivity, performance, and user experience.
As digital architecture and innovation partner for HM Courts & Tribunals Service (HMCTS), we’ve worked with service and technology leaders to explore AI’s role in enhancing productivity across the administrative support functions in the courts system, so staff can focus on the most complex tasks and accelerate backlog reduction.
We’ve developed a clear strategy to drive tangible value from AI in a responsible manner. Alongside this, we demonstrated how a generative AI knowledge management solution can unlock individual productivity, and enabled HMCTS to secure investment in AI solutions to help staff manage the increasing volume and complexity of casework.
We continue to work closely with HMCTS to put innovation into action, enhancing the services HMCTS delivers, enabling the court system to run more efficiently, and helping people access the services they need.
Embedding a responsible approach to AI
HMCTS is responsible for administering the criminal, civil, family courts, and tribunals in England and Wales as well as non-devolved tribunals in Scotland and Northern Ireland. Against a backdrop of high demand, challenging case backlogs, and rapid technology change, HMCTS recognised the need to identify, test, and scale innovative technology solutions – including AI – to continue delivering high quality public services.
However, the use of AI in the justice system raises important and complex legal, policy, and ethical challenges. Any application of AI must provide consistently accurate, rational, and unbiased outputs, be trusted and accepted across a large operational workforce, and crucially, respect the boundaries on judicial decision making.
Working closely with HMCTS and the judiciary, we developed a strategy for targeted and focussed AI adoption that enabled HMCTS to realise the benefits presented by AI while managing the risks posed by emerging technology. The Responsible AI approach was underpinned by nine principles that guide the use of AI in HMCTS to ensure it is appropriate, safe, and controlled. Our work has enabled HMCTS to take a risk-based approach to AI adoption, testing low-risk ideas for “back office” processes, before considering whether AI could be used in frontline services.
Working with PA, we sought to make innovation part of business as usual – to continually identify opportunities to apply cutting-edge technologies so HMCTS can deliver support services that improve the courts and tribunals for those who use them, and those who work in them.”
Given the risks posed by using AI in the administration of the justice system, putting the right ethical guardrails in place from day one was a critical strategic priority, so that future innovation would protect the integrity of the justice process.”
Using generative AI to unlock productivity gains
HMCTS identified effective knowledge management as a key challenge for frontline staff. Currently, staff have access to a wealth of operational procedures and guidance documents – meaning it can take tens of minutes to find relevant content, with staff often eventually resorting to colleagues for advice, impacting productivity of the whole team.
Our team of strategists, developers, data scientists, and AI ethicists worked with HMCTS to design, build, and pilot a generative AI knowledge retrieval assistant, working in collaboration with Microsoft to build the system using Microsoft Azure OpenAI Services. Staff can ask questions using natural language and the tool interrogates over 300 unstructured documents before returning a simple summary, accompanied by a citation to the source document.
Nitesh Soni, PA’s AI Engineering expert describes the impact of our work: “We’ve worked with HMCTS’ operational users to show generative AI provides access to the information they need so they can digest key information more quickly, ultimately increasing the pace of case administration in the criminal justice system.”
Improving the quality and speed of manual tasks
HMCTS processes over eight million paper forms every year, and, due to limitations with current automation systems, staff currently spend valuable time manually uploading and reviewing the forms. This increases the cost and time of case processing, diverting staff from higher-value activities.
We applied Intelligent Document Processing (IDP), which uses machine learning and computer vision technology, to automate the extraction and analysis of information from paper-based forms. We demonstrated that IDP processed forms more accurately, enabled a more intelligent review process, and automatically configured form structures. This all contributed to a reduction in the volume of exceptions, quicker case processing times, and reduced configuration costs.
Following testing, we developed a business case for large-scale adoption of the solution to express the size of the opportunity and implications of implementation, leading to funding being secured. “PA’s experience in developing government-standard business cases was instrumental in helping bring AI into our operational processes”, explained John Laverick.
Delivering swift access to justice for those who need it most
We continue to work closely with HMCTS to put innovation into action, enhancing the services HMCTS delivers, and enabling the administrative systems to run more smoothly and efficiently. For the public purse, the gains in productivity mean better value for money on behalf of taxpayers. For people waiting for life-changing decisions, HMCTS will be better placed to help and support those most vulnerable in society.
Explore more
