Guest Speaker: Distinguished Prof. LI Hao
Advanced Institute for Materials Research
Tohoku University
Professor Hao Li is a tenured Distinguished Professor at the Advanced Institute for Materials Research (WPI-AIMR), Tohoku University, Japan. He is the only Chinese scholar currently holding the title of Distinguished Professor in Japan and is among the youngest tenured full professors in Japanese university history. He received his Ph.D. from UT Austin under Prof. Graeme Henkelman and completed postdoctoral research at the Technical University of Denmark with Prof. Jens K. Nørskov..
His research focuses on AI-driven materials discovery, catalysis, all-solid-state batteries, and hydrogen-storage materials. He developed digital platforms like DigCat, DigBat, and DigHyd, which combine databases, machine learning, physics modeling, and automated experimentation into closed-loop frameworks to accelerate functional materials R&D.
Since 2014, Prof. Li has published over 400 peer-reviewed papers in top journals (such as Nature family journals, JACS, PNAS, and Angewandte Chemie), earned the MEXT Young Scientists’ Award along with ACS and AIChE honors, and been named a highly cited scientist. He is Founding Editor-in-Chief of AI Agent and co-Editor-in-Chief of AI for Materials.
He has led research projects securing over RMB 200 million in funding and delivered over 150 invited lectures. In 2025, he founded Suzhou MatSource Technology Co., Ltd., serving as Chairman and Chief Scientist to advance AI-enabled materials R&D and industrial deployment. The company has raised several hundred million RMB, and in 2026, he was selected for the Suzhou Industrial Park Major Leading Talent Program.
Abstract
This presentation introduces the Digital Materials Ecosystem, a closed-loop research framework centered on digital platforms for AI-driven materials discovery. With the rapid growth of experimental, computational, and literature data, the key challenge is transforming diverse data sources into reliable scientific knowledge, predictive models, and practical applications.
To address this, the ecosystem integrates large-scale databases, AI, physics-constrained models, automated experimentation, high-throughput computation, and expert feedback. At its core are digital platforms for energy materials including DigCat (catalysts), DigBat (batteries), and DigHyd (hydrogen storage) which serve as digital research infrastructures connecting data, mechanistic understanding, and performance metrics.
Building on these platforms, AI agents extract knowledge from scientific literature, learn from integrated data, predict promising materials, and propose synthesis strategies. Physics-based models are deeply integrated throughout the workflow to enhance prediction reliability and ensure candidates are scientifically meaningful and experimentally feasible. By connecting data generation, AI prediction, theoretical validation, and experimental feedback, the ecosystem replaces conventional trial-and-error exploration with an autonomous, continuously learning paradigm that substantially accelerates functional materials discovery.
