Ekin Dogus Cubuk

Liam Fedus, co-creator of ChatGPT and now co-founder of Periodic Labs, teams up with Ekin Dogus Cubuk, former head of materials science at Google DeepMind, to discuss their innovative approach to automating physical science discovery. They dive into the importance of integrating LLMs with experimental data to tackle complex problems like superconductivity and magnetism. Their lab focuses on real-world experimentation, emphasizing the necessity of iteration and learning from failures. With a unique hiring philosophy and a commitment to bridging ML and experimental cultures, they aim to revolutionize how we engage with hard sciences.

Ekin Dogus Cubuk, co-founder of Periodic Labs and former Google DeepMind researcher, dives into the exciting intersection of AI and materials discovery. He reveals how advancements like OpenAI's O1 model are helping to tackle AI's limitations in predicting beyond its training data. The conversation shifts to the importance of breakthrough discoveries versus incremental ones. Ekin also discusses the synergy between human insight and AI automation, and how Periodic focuses on automated labs to refine experimental hypotheses.

Ekin Dogus Cubuk, a research scientist at Google DeepMind, specializes in AI applications for materials discovery. He discusses the complexities of using AI in finding new materials for climate tech, highlighting that current methods still require substantial lab work and larger datasets. The conversation also explores how AI might enhance the understanding of material properties like optical and electric qualities, and the challenges faced due to inconsistent data. Dogus emphasizes the need for quality experimental data to make AI truly impactful in this field.