MIT’s Rafael Gómez-Bombarelli: Leveraging AI to Transform Material Science
From the laboratories of MIT, Associate Professor Rafael Gómez-Bombarelli has been leveraging the power of artificial intelligence (AI) to engineer new materials for over a decade. As AI continues to advance, Gómez-Bombarelli’s aspirations have expanded in tandem, leading him to believe that AI stands on the brink of revolutionizing science in ways previously unimaginable.
Recently appointed as the professor of materials science and engineering, Gómez-Bombarelli is committed to expediting this scientific revolution. “We are at a second turning point,” he asserts, indicating that we are on the cusp of a significant leap in the integration of AI into science, particularly in the realm of material structures, synthesis recipes, and language-processing abilities.
Blending Physics and AI for Material Innovation
Gómez-Bombarelli’s research is a unique amalgamation of physics-based simulations and AI techniques, including machine learning and generative AI. This approach has led to the discovery of innovative materials with applications in batteries, catalysts, plastics, and organic light-emitting diodes (OLEDs). His entrepreneurial spirit has also seen him co-found several companies, including Lila Sciences, which aims to create a scientific superintelligence platform for the life sciences, chemical, and materials science industries.
His ultimate goal is to streamline and enhance the productivity of scientific research in the future. Gómez-Bombarelli views AI as a powerful tool that can propel science towards a better future, faster than ever before.
Gómez-Bombarelli’s Journey: From Lab Experiments to AI Simulations
Born and raised in Spain, Gómez-Bombarelli has been passionate about science from a young age. His academic journey in chemistry commenced at the University of Salamanca, where he completed his undergraduate studies and doctoral thesis on DNA-damaging chemicals. The turning point came midway through his doctoral research when he developed an interest in simulations and computer science.
“I started simulating the same chemical reactions that I had measured in the lab. I like the way programming organizes the brain; it felt like a natural way to organize one’s thinking,” Gómez-Bombarelli explains.
His postdoctoral research in Scotland led him to explore quantum effects in biology, which subsequently connected him with Alán Aspuru-Guzik, a chemistry professor at Harvard University. This collaboration marked the beginning of Gómez-Bombarelli’s journey into deep learning and AI in science.
He also began to automate parts of molecular simulations to conduct more high-throughput experiments, leading to the discovery of hundreds of promising materials.
Real-World Applications and the Future of AI in Science
Gómez-Bombarelli is not only focused on AI and simulations but also on the real-world applications of the materials he develops. His lab collaborates closely with companies and organizations such as MIT’s Industrial Liaison Program to understand the material needs of the private sector and the practical hurdles to commercial development.
As AI continues to gain momentum, Gómez-Bombarelli has witnessed the maturation of the field. Tech giants like Meta, Microsoft, and Google’s DeepMind are now regularly running physics-based simulations akin to what he was working on back in 2016. This proliferation of AI across scientific domains underscores the crucial tipping point that Gómez-Bombarelli believes we are at.
“People think in natural language, we write papers in natural language, and it turns out that these large language models that do natural language have opened up the possibility of accelerating science,” he says.
Ultimately, Gómez-Bombarelli’s work exemplifies the power of AI in advancing scientific discovery. His relentless pursuit of innovation serves as an inspiration, proving that the integration of AI into scientific research is not a distant dream but a rapidly approaching reality.
For more about Gómez-Bombarelli and his work, click Here.
