ICN2 played a major role in this prestigious conference, which is becoming one of the leading events in Europe at the intersection of artificial intelligence and materials design.
From 8 to 10 April, the second edition of Artificial Intelligence for Advanced Materials (AI4AM2025) was held in San Sebastián, Spain. This international conference brought together a wide range of artificial intelligence (AI) and materials science experts from industry, research and academia. Through various talks and panel discussions, the event highlighted how AI is paving the way for a new era in the design and production of materials that could play a key role in areas such as electronics, energy, biomedicine and computing.
ICN2 played a prominent role in this edition of the event, being one of the main organisers alongside institutions and entities like the Phantoms Foundation, the Donostia International Physics Center (DIPC), the National University of Singapore (NUS) and Constructor Tech. CSIC Prof. Pablo Ordejón and ICREA Prof. Stephan Roche represented the Institute as members of the International Scientific Committee and the Organising Committee respectively.
The conference was attended by representatives from prestigious institutions and major technology companies such as UC Berkeley or NVIDIA. Several ICN2 researchers took part as speakers through talks and lectures that explored how machine learning and AI models are becoming powerful tools to design new materials and better understand their properties:
- ICREA Prof. Stephan Roche: "Intelligent Workflows for Designing Innovative Advanced Materials"
- Andrei Voicu Tomut: "Demonstration of Scientific Workflow Applied to Disordered Quantum Materials"
- Dr Jose Hugo García: "Equivariant AI-based models for accurate electronic Hamiltonians"
- Dr Emigdio Chavez Angel: "ML-driven Thermal Sensing Using FTIR Spectroscopy"
In addition, Prof. Stephan Roche was one of the programme chairs and moderated the roundtable: Towards a Unified AI Framework – Methodological and Infrastructural Perspectives for the EU. Some ICN2 researchers also participated as speakers in the PhD sessions or presented posters on their research lines:
- Onurcan Kaya: "Revealing Structure-Property Relationships in Amorphous Boron Nitride Using Machine-Learned Potentials"
- Sara Navarro: "Developing Accurate Exchange-Correlation Functionals through Physics-Informed Machine Learning"
- Jaime Garrido. Poster: "Studying 2D magnetic materials with high-throughput automated workflows from Density Functional Theory”
- Thomas J. Galvani. Poster: "Dielectric properties in models of amorphous Boron Nitride”
- Dorye L. Esteras. Poster: "Towards automatic workflows to accelerate the discovery of quantum materials"
- Andrei Voicu Tomut. Poster: "Machine Learning-Driven Hamiltonian Matrix Prediction: Equivariant vs. Non-Equivariant Models”
The role of MaX
"MaX", a European project in which ICN2 is a partner, participated as a sponsor and exhibitor. Through their information stand, the MaX team had the opportunity to showcase their work and explain how the project is dedicated to using High-Performance Computing (HPC) to model, simulate, discover and design materials.
At ICN2, we celebrate the success of this edition of AI4AM2025 and are confident that the event will continue to grow and strengthen in the years to come.
(This initiative is supported by the Severo Ochoa Centres of Excellence programme, Grant CEX2021-001214-S, funded by MICIU/AEI/10.13039/501100011033).
