AI for court and case file summarization

Prosecutors in England and Wales review recorded video interviews as part of case preparation. Beam Notes lets them upload a recording and receive a time stamped transcript and a structured summary (key details such as names and dates, and a chronology of events) built on templates that the CPS co designed with the supplier. Prosecutors must still watch the original video, every summary is reviewed before anything is moved into case systems, users rate summaries in the app, and the tool gives no advice on charging or case outcomes. The transparency record says it will be used across all 14 CPS areas on about 11,000 cases a year, with data kept by the supplier for up to 30 days.

The City of Amsterdam's legal department keeps an internal case library of its advice on objections (bezwaren) against municipal decisions. The city has registered a tool, built in house, that would use an OpenAI GPT model (gpt-3.5-turbo or gpt-4) through Azure to write a summary of the core of each existing advice, shown first in the library, so lawyers handling a new objection can judge more quickly whether an earlier advice is relevant.

Users can report errors in a summary, which are then corrected; the register says summaries will be marked as made with generative AI and checked by sampling. The register entry is marked "In gebruik" (in use) with a start date of July 2022, but its method section says the algorithm still has to be developed and that the prompting approach is yet to be decided, so the tool is recorded here as announced. No outcome figures are published.

The UAE Ministry of Justice showed five specialized AI agents that work with its electronic court systems and support judges through a seven stage case review, five of which have AI support. One agent classifies digital case documents and extracts key information, one links facts and evidence and identifies the legal issues, one compiles legislation and precedents, one sets out the legal options for the judge and one drafts a non binding preliminary judgment that the judge reviews, amends and approves. The ministry says that by deploying these assistants it cut the review of case files from several days to a few hours, and that deployment will expand after further evaluation and integration with partner entities.

Brazil's Federal Supreme Court runs Maria, a generative AI platform inside its STF Digital environment that supports court staff in analyzing cases and producing documents. It started by generating headnotes (ementas) in the National Council of Justice's standard format, case reports (relatórios) for extraordinary appeals and questionnaires for initial petitions in constitutional complaints (reclamações), and has been extended to more case classes, grammatical review and a unified search of related precedents.

The court stresses that Maria never acts autonomously: staff review, adapt and decide whether to use each report or headnote. It builds on earlier tools, Victor (2018) for triaging extraordinary appeals and VitorIA (2023) for grouping similar cases. The court is deploying open source language models on servers that will soon be available in its own data center. No outcome figures are given in the article.

The Civil Division lists a pre deployment initiative to apply AI to synthesize case records, summarize expert reports and depositions, identify duplicate claims, flag inconsistencies between records and plaintiff claims, and analyze settlement and damages databases for outlier trends. The stated aim is faster evidence review, fewer duplicate or weak claims going forward, and attorney time saved, building on the department's existing medical record review systems, HHS and VA data integration, and its Relativity and CORA settlement and damages databases. No system is live yet and no outcome is published.

The Executive Office for Immigration Review, which runs the US immigration courts, lists a pre deployment use case for summarizing court filings. Filings can run to hundreds of pages and are often poorly organized; the planned tool would summarize their contents with references to the source of the information, tab and label submission types, point adjudicators and legal support staff to where relevant content sits in the record, and summarize case law for training material. The stated aim is to let adjudicators spend their time on legal analysis and conclusions.

The Home Office piloted an Asylum Case Summarisation tool that uses a large language model to summarize asylum interview transcripts, so decision makers can get back up to speed on a case faster. Two eight week phases ran in 2024 in two decision making units, with a test group using the tool and a comparison group working as usual. The department's own evaluation, published in April 2025, found that the test group reviewed transcripts 23 minutes quicker per case, a 32% time saving, with no visible effect on decision quality in a dip sample. Summaries stayed an aid, and 9% of them were removed from the pilot as inaccurate or incomplete.

Prometea is an artificial intelligence system built inside Argentina's Ministerio Público Fiscal de la Ciudad Autónoma de Buenos Aires (the public prosecutor's office of the City of Buenos Aires), in operation since 2017. It uses natural language interaction and supervised machine learning to read case documentation, draft standard documents such as reports, decrees and rulings, and predict the likely outcome of a case for staff to review. The prosecutor's office reports large reductions in drafting time in real cases, and Prometea now also runs in institutions in the provinces of Mendoza, Chaco, Corrientes and Santa Fe.

In September 2026 the Abu Dhabi Judicial Department launched the first phase of its Judicial AI Platform, built with the Department of Government Enablement. Phase one is the Judicial Assistant, which helps prosecutors and criminal court judges analyze specific types of cases, recommend decisions and draft judgments that cite legislation and judicial precedent. Final decisions stay with the judicial authority member and the AI recommendations are not binding. The launch is the first step of an 18 month roadmap that is to add real time translation, automated hearing records and tools to detect forged documents, and to extend the platform to civil and other ADJD systems.

Oman's Public Prosecution, working with Rahal, piloted an AI system that analyzes judicial documents, generates questions for the parties to a case and drafts decisions on whether a case is dismissed or referred. Sensitive data is processed locally. The pilot was one of several announced in December 2025 by the Ministry of Transport, Communications and Information Technology as low risk public private projects, each supported by the implementing company. The ministry reports an accuracy rate above 90 percent and does not say how accuracy was measured.