We are very excited to announce the official launch of the alignAI project: value-ALIGNed socio-technical systems using large-language models (LLMs). Supported by the European Union’s Marie Skłodowska-Curie Action (MSCA) funding, alignAI has been awarded over €3.5 million to fund doctoral research across a network of esteemed institutions. The TUM IEAI will lead and coordinate this four-year-long project.
This project will train the next generation of researchers to develop LLMs that align with critical human values, ensuring these advanced technologies are ethically sound and socially responsible.
13 doctoral candidates will be hosted by EU partners at:
- Technical University of Munich (TUM)
- Technical University of Eindhoven (TU/e)
- Technical University of Denmark (DTU)
- Region Hovedstaden (RegionH)
- Technion – Israel Institute of Technology (Technion)
Two Swiss research institutions have also joined alignAI, with additional funding supporting four doctoral candidates at:
- École Polytechnique Fédérale de Lausanne (EPFL)
- IDIAP Research Institute
To foster real-world application and hands-on learning, alignAI has partnered with key industry organisations that will offer secondment opportunities, including:
- DPG Media
- EHS Médias
- Datacation
- Fujitsu Germany
As LLMs become increasingly integrated into society, the alignAI project will focus on essential areas such as explainability and fairness, particularly in applications related to education, mental health, and media. By bringing together expertise from social sciences, humanities, and technical disciplines, alignAI aims to address complex ethical challenges while creating tools and guidelines for AI systems that reflect and uphold human values.
The alignAI project will substantially impact AI research, ensuring that the development and deployment of LLMs contribute positively to society. The project represents an essential step toward responsible AI advancements for a brighter, more inclusive future, in line with the core mission of the TUM Institute for Ethics in AI.
For more details on the project and to stay up to date with our research activities, please check out the alignAI website and follow the project on social media: LinkedIn or X.