Five Berlin institutions unite for AI materials research

Five of Berlin's biggest research institutions just made it official: they're teaming up to share data and push materials science forward with the help of artificial intelligence.

Humboldt-Universität zu Berlin (home to the NFDI consortium FAIRmat), Helmholtz-Zentrum Berlin (HZB, or the Helmholtz Centre for Materials and Energy), the Leibniz-Institut für Kristallzüchtung (IKZ, the Leibniz Institute for Crystal Growth), the Paul-Drude-Institut für Festkörperelektronik (PDI, the Paul Drude Institute for Solid State Electronics) and the Bundesanstalt für Materialforschung und -prüfung (BAM, the Federal Institute for Materials Research and Testing) signed a Memorandum of Understanding to build a long-term collaboration around research data, data infrastructure and AI.

The idea is simple but ambitious: make research data accessible across institutional lines, agree on shared data and metadata standards, and get more out of AI methods in materials research. A big piece of the puzzle is the further development of NOMAD, the open research data platform built for materials science.

The signing took place at the FAIRmat–HZB Symposium "Digital Future of Materials Science," which opened with a keynote from Giorgio Rossi of the University of Milan and NFFA on why shared research infrastructures matter for data-driven science. FAIRmat spokesperson Claudia Draxl walked through how NOMAD is preparing the field for new data-driven approaches, while HZB's Bernd Rech spoke about what AI means for science and technology going forward.

HU Berlin president Julia von Blumenthal called the agreement a step that will speed up scientific progress and help keep Berlin at the forefront of AI in materials science. The symposium wrapped with a panel on the cultural shift needed to make the most of research data and AI, a reminder that the technology is only half the story.