Innovation That Matters

| Photo source © Julia Filirovska from Pexels via

A generative AI search engine for advanced materials


The platform can be used to develop new carbon-capturing materials to help tackle climate change

Spotted: Tackling climate change will rely on a range of new technologies and materials, but testing these out can be time-consuming, expensive, and wasteful when things don’t go to plan. What if we could use generative AI to design the specific material we need from the outset?

This is exactly what startup CuspAI set out to do, with its search engine that allows users to request and evaluate the properties of existing and new materials on demand. For example, users can request a material that selectively binds carbon dioxide under specified conditions. The platform then uses generative AI, deep learning, and molecular simulation to generate, evaluate, and optimise potential molecular structures that meet those exact criteria.

Although Cusp’s tech could theoretically be used to produce any type of material, the company is focusing initially on materials that could help capture carbon emissions. This is important, as current direct air capture (DAC) technology can be inefficient and expensive and for the technology to grow, DAC companies need to get their costs down to $100 per tonne or less.

Cusp recently emerged from stealth with a $30 million seed funding round led by Hoxton, with participation from Basis Set Ventures and Lightspeed Venture Partners, as well as angel investors from Google Deepmind Ghissassi and Dorothy Chou. The startup has also partnered with the Fundamental AI Research (FAIR) team at Meta to further its search for new novel DAC sorbent materials.

Written By: Lisa Magloff



Download PDF