FOR 5906 - TP 5: Use of Artificial Intelligence in Civil Proceedings

Basic data for this project

Type of projectSubproject in DFG-joint project hosted at University of Münster
Duration at the University of Münster01/05/2026 - 30/04/2030 | 1st Funding period

Description

Civil procedural law plays a key role in fulfilling core legal functions, such as balancing private interests, ensuring legal peace and protecting fundamental rights. This sub-project aims to investigate whether these functions can be fulfilled to the same or even a greater extent if humans relinquish direct responsibility and decisions are partially or completely taken over by AI. Firstly, we would like to analyse whether and why human decision-making (anthropocentrism) is necessary in civil proceedings. Unlike in criminal procedural law, we assume that the use of AI is not only permissible if the character of a human decision is preserved. The challenge in civil proceedings is therefore to determine the situations in which human decision-making is essential and those in which the use of AI for judicial tasks would be desirable. We are taking a positive approach and would like to consider whether AI can improve the quality and acceptance of decisions in certain areas. However, the familiar problems associated with the use of AI also arise in civil proceedings. For example, with AI, it must also be ensured that cognitive biases are avoided, and that human biases are not reinforced. Even if AI is not intended to make the final decision, there is a risk that the responsible judges will rely too heavily on it (automation bias). This issue is particularly more complex in civil proceedings than has usually been discussed to date. It is not just about the risk of judges recklessly following the AI's suggestions. It is also possible that they will reject the AI's proposal too quickly if it contradicts their subjective opinion, despite being based on objective facts. We would like to analyze these problems and the question of acceptance comprehensively in empirical studies. To this end, experimental vignette studies are to be carried out. Secondly, experimental economic studies involving judges, among others, are planned. The findings from these studies will then provide the legal and factual basis for a differentiated legal analysis.

KeywordsKünstliche Intelligenz; Privatrecht
DFG-Gepris-IDhttps://gepris.dfg.de/gepris/projekt/574908975
Funding identifierHE 3664/2-1; GO 4420/1-1; RU 2946/1-1 | DFG project number: 548899139
Funder / funding scheme
  • DFG - Research Unit (FOR)

Project management at the University of Münster

Göhsl, Jan-Frederick
Associate professor of Civil law, commercial law and legal issues of digitalisation (Prof. Göhsl)
Heiderhoff, Bettina
Professor of Private International Law, International Civil Procedure and German Private Law
Rüsing, Christian
Professor of Private International Law, International Civil Procedure and German Private Law

Applicants from the University of Münster

Göhsl, Jan-Frederick
Associate professor of Civil law, commercial law and legal issues of digitalisation (Prof. Göhsl)
Heiderhoff, Bettina
Professor of Private International Law, International Civil Procedure and German Private Law
Rüsing, Christian
Professor of Private International Law, International Civil Procedure and German Private Law

Project partners outside the University of Münster

  • University of BonnGermany