AI is being explored as a potential game-changer in the field of materials research, particularly for applications in climate technology, where new materials like batteries, semiconductors, and magnets are in demand. A report by David Rotman highlights that while AI presents opportunities for innovation in discovering novel materials, it must still demonstrate its ability to create practical and effective solutions.
The challenge in material invention is significant and often lengthy. For example, the first synthetic plastic was created in 1907, but it took several decades for companies to develop a broad range of plastic products. Materials science has seen limited breakthroughs in recent years, with notable exceptions like lithium-ion batteries. Rotman suggests that AI’s potential in this domain could represent a major shift, prompting many companies to explore its capabilities.
One such company, Lila Sciences, based in Cambridge, Massachusetts, is using AI to enhance the discovery of new materials. The company’s approach involves training AI models on current scientific literature and integrating these models with automated laboratories to learn from experimental results. The aim is to accelerate the process of inventing and testing materials, allowing researchers to explore options that might otherwise be overlooked.
At a recent MIT Technology Review event, Rafael Gómez-Bombarelli, co-founder of Lila Sciences, spoke about the company’s endeavors in AI-driven material discovery. He noted that while the field has not yet experienced a significant breakthrough, the potential remains. This raises questions about the future role of AI in materials science: Can it indeed revolutionize the field, and what would such advancements entail?
Source: https://www.technologyreview.com/2025/12/18/1130102/ai-materials-discovery/