In a nondescript laboratory on the outskirts of Cambridge, researchers have spent decades testing combinations of elements in the hope of finding a material that could slash the energy demands of the next generation of computer chips. That slow grind is what CuspAI wants to change.
The British artificial intelligence startup announced on Monday that it had closed a $450m series B funding round, lifting its valuation to $2.6bn. The round drew capital from Bezos Expeditions, the UK government's Sovereign AI Venture Fund, Kleiner Perkins, NEA, Glade Brook Capital Partners, Lux Capital, AMD Ventures, Invest-NL and the venture capitalist John Doerr.
This is not another story of speculative hype around large language models. CuspAI builds tools that hunt for physical substances that do not yet exist, substances capable of making semiconductors more efficient, batteries longer-lasting and clean-energy devices less dependent on scarce metals such as iridium and ruthenium. The company says its approach can cut what used to be a ten-year research cycle to something far shorter.
The funding coincides with the launch of CuspAI's AI Materials Foundry, a network of more than 45 partners that pools data, laboratories, computing power and expertise. Participants include NVIDIA, Meta, Samsung, Hyundai Motor Group, Applied Materials and Tokyo Electron. The Foundry is designed to turn the startup's algorithms into a shared infrastructure for industrial discovery.
Dr Chad Edwards and Prof Max Welling, the co-founders, put the stakes plainly in the company announcement.
If we don’t make progress fast, the next 50 years of industrial progress will be constrained by a single challenge: the world needs materials that don’t yet exist.
Josh Coyne, a partner at Kleiner Perkins, framed the opportunity in similar terms.
