Materials Science and AI: Promises, Pitfalls, and Paths Forward

Prasanna Balachandran
UVA Material Science and Engineering

Time: 2025-11-19, 12:00 - 13:00 ET
Location: Rice 540 and Zoom

Abstract In a 2008 report, the National Academy of Engineering identified 14 Grand Challenges for engineering in the 21st century. New materials are essential for addressing at least 10 to 12 of these challenges. However, the traditional approach to materials discovery is slow, which poses a barrier to meeting urgent technological needs. A current example of this challenge is the rapid expansion of data centers in the Commonwealth of Virginia. Sustaining growth in this sector requires innovative materials science solutions, including the development of new nuclear materials, making accelerated materials discovery pipelines essential. This talk will discuss the status of artificial intelligence (AI) and machine learning (ML) in materials design and discovery based on recent literature, highlighting both successes and limitations. A key takeaway is that effectively implementing AI and ML in materials science requires more than advanced algorithms; it necessitates the thoughtful integration of physics-based constraints and domain-specific knowledge within a framework grounded in physics.

Bio: Prasanna Balachandran is the Heinz and Doris Wilsdorf Research Associate Professor of Materials Science and Engineering at the University of Virginia (UVA). Prior to joining UVA in December of 2017, he was a postdoctoral research associate in the Los Alamos National Laboratory and Drexel University. He received his PhD in Materials Science and Engineering from the Iowa State University in 2011. His Materials Informatics Group develops and applies computational materials science methods, including AI/ML, to accelerate the search and discovery of new materials with targeted properties.