Accelerate insight with AI-powered search and knowledge retrieval
Read the full detail of action 2.1.2 on GOV.UK. (opens in a new tab)
Year 1 progress
In Year 1, significant progress has been made in accelerating insight through AI-powered search and knowledge retrieval. Semantic search capabilities are already live, scaled and delivering value, enabling faster access to relevant information across operational data.
Year 2 priorities
In Year 2, we will focus on progressing AI-powered search and knowledge retrieval tools, expanding use cases, strengthening integration across systems, and using proven models to support insight generation consistently across the justice system.
Case studies (2)
Delius Contact Log Semantic SearchATRS — record published on the Algorithmic Transparency Recording Standard (opens in a new tab)Scale
This AI-powered semantic search tool is helping probation officers find the information they need more quickly and accurately, reducing the time spent searching for guidance and information.
Probation officers work with large volumes of information, including case records and operational guidance. Traditional keyword searches can make it difficult to locate the most relevant information, leading to missed information when authors have used different language and time-consuming “search flurries” to find what they need.
To address this, Justice Digital, Data & Science (JDDS) introduced a semantic search capability within the Probation Digital System. Powered by a Large Language Model (LLM), the tool understands context, language variation and abbreviations, enabling it to return more accurate results from the outset. Unlike basic keyword search, it can interpret synonyms, abbreviations, and misspellings to provide results that match user intent.
This helps practitioners access critical information more efficiently, supporting better risk management and more effective rehabilitation outcomes.
LauriumScale
This AI-enabled software package helps analysts and policymakers unlock insights from unstructured data by extracting structured information from free-text records.
Government services generate large volumes of free-text data through daily interactions with the public. While rich in insight on service effectiveness, emerging needs and operational challenges, this information is often locked in unstructured formats across systems and underused in decision-making.
To address this, Justice Digital, Data & Science (JDDS) developed Laurium through the MOJ's Better Outcomes through Linked Data (BOLD) (opens in a new tab) programme, enabling case notes to be analysed at scale. Since September 2025, Laurium has been rolled out across multiple use cases, including parental imprisonment, Community Sentence Treatment Requirements, children on remand, prison safety, and drug recovery wings - strengthening the frontline evidence base and supporting policy development.
In recognition of its innovation, Laurium won the Analysis in Government Award (Innovative Methods) in January 2026.
Other departments are now adopting Laurium for analytical use cases. We plan to continue expanding use cases and adoption across government.
