Conferences and seminars

Conference: Explainability for legal decisions in the era of AI

Illustrasjonsbilde
Photo: COLOURBOX

The opening conference for the LEXplain project will explore the relationship between legal and computational explainability. The conference will take place on 15 December at the Faculty of Law.

AI technology used in the context of legal decision making, challenges several rule-of law ideals such as transparency in reasoning, accountability and relevancy of the explanation to the case at hand. In short, the use of AI for legal decision-making challenges lawʼs legitimacy. LEXplain investigates how AI technology can be adapted to case processing and decision making in a way that ensures compliance with the legal obligation to provide justificatory reasons. The conference will be organized around three thematic panels:

  1. Artificial Legal Reasoning
  2. Administrative Practice and Case Based Reasoning Systems
  3. Large Language Models in the World of Law 

Conference opening

08:30-09:00 Registration and coffee 

Panel 1: Artificial Legal Reasoning

09:00 - 11:00

Chair: Ragna Aarli

  • LEXplain: Researching the law and AI interface, Professor and PI LEXplain Henrik Palmer Olsen, University of Bergen
  • Artificial Legal Reasoning: Potentials and barriers, Assistant Professor Jennifer Raso, McGill University  
  • What is a good (enough) reason for a legal decision? Professor Mathilde Cohen, CNRS

Questions and discussions. 

11:00 - 11:30: Coffee break

Panel 2: Administrative Practice and Case Based Reasoning Systems

11:30 - 13:00 

Chair: Synne Sæther Mæhle 

  • Automatic for the people: The new Norwegian Public Administrative Act (forvaltningsloven 2025) has several provisions regarding automatic decision making. How will this change public administration, and are the principles of rule of law sufficiently safeguarded? Are there loopholes in the legislation that are overlooked? Professor Karl Harald Søvig, University of Bergen
  • Case Based Reasoning systems for decision support in public administration: How to build CBR systems to assist adherence to administrative practice, Postdoc Paul Cosma, University of Copenhagen
  • Panel discussion: How far have we come with AI as decision support for caseworkers in public mass administration that requires the use of discretion? Carine Røkenes (Head of Section at the Tax Appeals Board), Bjørn Eriksen (Senior Advisor at the Norwegian Tax Administration), Kaja Vollan Amundsen (Delivery Manager for Disability Benefits, Norwegian Directorate of Labour and Welfare) and Robindra Prabhu (data scientist, Norwegian Directorate of Labour and Welfare) in dialogue with Ragna Aarli

13:00 - 14:00: Lunch

Panel 3: Large language models in the world of law

14:00 - 16:00 

Chair: Thomas Hildebrandt

  • Automated legal information retrieval and analysis: state of the art and evaluation metrics, Professor Henrik Palmer Olsen
  • Computational approaches to assessment of case similarity, Docent Johan Lindholm, University of Umeå
  • How legality challenges the use of AI in public administration, Associate professor Aysel Kucuksu, University of Copenhagen 

16:00 - 16:30 Closing discussions

Contact person

Ingrid Brice Müftüoglu