woman with camera

Artificial intelligence can help courts improve efficiency and quality

Published: 25th May 2022 Last updated: 25th May 2022

In certain court cases a computer can autonomously generate draft decisions, and in almost all court cases it can offer valuable support to courts and court clerks. Yet even though artificial intelligence can improve efficiency and quality, the added value that human judges bring to the process of administering justice remains indisputable. This is the outcome of Manuella van der Put’s PhD research, which she will defend at Tilburg University on Wednesday, May 25, 2022.

Work pressure at Dutch courts is perceived to be high and rising, and this has a ripple effect on the entire legal system. Artificial intelligence (AI) might hold the key to reducing the pressure of work and to maintaining and improving the quality of the judicial system. For example, AI is capable of processing large amounts of information fast and of recognizing and describing patterns – all features of the work of courts and court clerks. And to an extent, AI is already doing this.

In her PhD research, Manuella van der Put, senior judge at the East Brabant court, specifically addressed the process of judicial decision-making to explore the potential of AI for the judicial system. To that end, she reviewed literature and met with multidisciplinary focus groups of legal and AI experts. After thorough analysis and using specific design and quality criteria, she had an AI system designed to decide cases involving minor traffic violations, the so-termed Mulder cases. She subsequently tested and evaluated the system. It is awaiting implementation.

Useful support for courts and court clerks

There are various stages in the process of judicial decision-making, Van der Put has found, in which AI can play a role. In certain court cases (as yet only uncomplicated ones), the computer can decide autonomously, and in almost all court cases it can offer valuable support to increase efficiency and quality even further.

For example, in the relatively straightforward Mulder cases, the system was capable of identifying the facts of a case, of verifying information such as the appeal period and the security deposit (whether a fine has been paid), of analysing a case by linking the information to similar cases in the database, and of suggesting a decision based on the most frequent outcome in similar cases. The information thus generated saved time and broadened understanding of case law. If the system evolves, it should also be capable of justifying the decision it reaches on the basis of reasons given in earlier cases and of drafting a judgement using text blocks and formats.


Human judges are indispensable

The system has proved incapable as yet of independently taking into account relevant statutory provisions and case law in the final decision in individual cases, as it did of legally qualifying facts. Judges and artificial intelligence need each other, Van der Put concludes. Human judges clearly add value to the administration of justice, but AI can play a major role in improving efficiency and quality.


The judiciary should embrace AI, Van der Put contends. It is and will continue to be important to explore how AI can contribute to meeting the challenge of meeting the needs of both individual citizens and society.

PhD Defense

Manuella van der Put defends her PhD thesis in the Auditorium of Tilburg University on Wednesday, May 25, 2022, at 16:30 hrs. Title of the doctoral thesis: Kunstmatige intelligentie bij rechterlijke oordeelsvorming: theoretische analyse en praktische implementatie (artificial intelligence in judicial decision-making: theoretical analysis and practical implementation). Supervisors: Professor J.E.J. Prins, Professor F.J. Bex, Professor J.A. van den Born. The defence can be followed through a livestream.

Note to editors

For more information about the research, please contact Manuella van der Put through cvrb@rechtspraak.nl. A review copy of her PhD thesis is available on request at persvoorlichters@tilburguniversity.edu.