Molecular Universe

10^11 is both the number of humans ever lived on Earth and the number of small molecules that could be used for batteries. For our Human Universe, all data in the form of books, videos, social media, and others generated over the past 200,000 years are being used to develop Artificial General Intelligence (AGI). Similarly, we want to map all fundamental physical and chemical properties of small molecules to build a Molecular Universe (AI4Science). This will help train an AI Agent to accelerate material discovery for Li-Metal and Li-ion batteries and accelerate the transition to electric transportation.

1011

# of people ever lived on earth
All knowledge, books, social network profiles, everything about everything and everyone

AGI

# of small molecules that could be used for batteries
All properties about small molecules

AI4SCIENCE

Our AI4Science effort begins with mapping the universe of small molecules

Here we leverage the latest compute hardware and computational chemistry and physics software to build the world’s largest molecular property database. Then we train advanced AI models with this molecular property database, literature and our team of world-class domain experts to both screen and generate new molecules.

These new molecules are then synthesized in our Electrolyte Foundry and formulated in electrolytes and tested and validated in actual cells (from lab samples to A-samples to B-samples and beyond).

How it works

All small molecular properties 10^11
All battery, material science and physicochemistry related papers and books 10^6
All SES internal data
AI/ML model development and training
Recommend new molecules 10^3
Electrolyte Foundry molecule synthesis and electrolyte formulation
Battery testing and validation (lab, A-samples, B-samples, and beyond)

Research

“Electrolytes for advanced batteries have been determined by trial and error. Up to the present there has never been a viable electrolyte solvent, salt or additive first identified by a computer. This has to change! The Molecular Universe project promises to actually find viable and improved electrolyte components. I am very excited by this project!”

Dr. Jeff Dahn
Professor, Canada Research Chair, and NSERC/Tesla Canada Industrial Research Chair at Dalhousie University

“Molecular design for better electrolytes is a test case we hope to showcase the power of open-source database for accelerating AI for Science.”

Dr. Shirley Meng
Professor of Molecular Engineering at the University of Chicago and Chief Scientist for Energy Storage Science at Argonne National Laboratory

“Since decades electrolytes are the most underrated battery component in lithium-ion batteries and beyond battery chemistries, both in academia and in industry. The molecular universe initiative is the most serious and promising attempt to fundamentally change this awkward situation.”

Dr. Martin Winter
Professor and Scientific Leader of MEET Battery Research Center at the University of Münster and of Helmholtz Institute Münster of Forschungszentrum Jülich, Member of National Academy of Science and Engineering

“Artificial intelligence (AI) is beginning to have a transformative impact on drug discovery and is poised to do the same for materials discovery. Molecular Universe's machine learning tools, comprehensive database, and extensive computing resources will enable the material science community to fully harness the power of AI for the betterment of humankind.”

Dr. James J. Collins
Professor of Medical Engineering & Science at MIT and Member of National Academy of Engineering, National Academy of Medicine, and National Academy of Sciences

Industry

“Energy efficiency and minimal environmental impact are crucial in advancing computational workflows. As part of its collaboration with Crusoe, SES AI is using NVIDIA’s GPUs and AI technology to harness powerful compute capabilities that will help unlock new potential in battery innovation and energy storage solutions.”

Matt McGrigg
Director of Global Business Development for Cloud Partners, NVIDIA

“Crusoe is dedicated to aligning the future of computing with the future of the climate. By providing a sustainable and cost-effective AI cloud platform, we are supporting SES AI’s vision to map the molecular universe and create ground-breaking AI models that could accelerate a sustainable future. This collaboration showcases how advanced computing can drive scientific progress while minimizing environmental impact.”

Chase Lochmiller
CEO, Crusoe

“Supermicro’s high-performance computing solutions are designed to support intensive workloads and innovative projects like SES AI’s initiative in AI for Science. We are excited to contribute to this effort with our advanced and energy-efficient hardware.”

Cenly Chen
Chief Growth Officer, Supermicro

Stay in the know

By clicking “Subscribe”, you are agreeing to our privacy policy.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.