Research and Development Cooperation Project
This interdisciplinary research project aims to develop an AI-powered system capable of identifying and correcting potentially unlawful clauses in German employment contracts. Leveraging recent advancements in Large Language Models (LLMs), the system will assist users—both employees and employers—by automatically analyzing uploaded contracts, highlighting problematic clauses, and suggesting legally sound alternatives.
The project combines expertise in Natural Language Processing (NLP) from the Technical University of Munich (TUM) with practical legal experience from the labor law team at SYLVENSTEIN Rechtsanwälte. A scientific advisory board, including Prof. Dr. Hans Christoph Grigoleit, ensures strong academic grounding in legal theory.
Objectives:
Automate the legal review of employment contract clauses using LLMs.
Generate context-aware correction suggestions based on legal best practices.
Ensure up-to-date legal compliance through modular prompt-based adaptation.
Evaluate the tool’s practical utility through real-world testing and continuous legal supervision.
Project Partners:
TUM – sebis Chair (Prof. Dr. Florian Matthes): Responsible for AI development and system architecture.
SYLVENSTEIN Rechtsanwälte: Legal expertise, annotation of contracts, and practical validation.
This project is funded by the Bavarian Research Foundation (Bayerischen Transformations- und Forschungsstiftung).
