CuspAI, an AI materials science company founded in 2024 by Professor Max Welling and Dr. Chad Edwards, has collaborated with Kemira, a leader in sustainable chemical solutions, to design new materials for PFAS remediation. This partnership represents the first commercial application of end-to-end generative AI for designing new materials for this purpose. The project focused on three priority PFAS molecules: GenX, PFBS, and PFOS, aiming to remove them from water at trace concentrations using stable, sustainable, and manufacturable chemistry. CuspAI's platform explored approximately 300 trillion possible material structures and delivered over 5,000 novel designs, which were then narrowed down to about 20 priority candidates in just six months, demonstrating unprecedented speed and scale in discovery.
The newly designed materials are metal-organic frameworks (MOFs), a class of porous crystalline materials whose structure and chemistry can be precisely tuned for specific filtration and adsorption applications. The project delivered full property data for the selected designs. This initiative was driven by tightening regulations on PFAS globally, including US EPA maximum contaminant limits, and the limitations of current leading remediation technology, granular activated carbon (GAC). Kemira, already active in activated carbon regeneration, sought more selective and longer-lasting alternatives through AI-driven design.
The project has now entered its next phase of further development and testing. Kemira and CuspAI are also scoping additional programs across other material classes under their partnership framework. CuspAI, ranked as a top AI startup, has raised over $200 million in Series A and Seed funding from investors like NEA, Temasek, NVIDIA, Samsung, and Hyundai, with advisors including Martin van den Brink, Geoffrey Hinton, and Yann LeCun. The success of this collaboration highlights the potential of generative AI to compress discovery timelines and create significant impact in addressing pressing environmental challenges.