Nvidia, Meta back CuspAI's chip-materials consortium with $450M

CuspAI, a British artificial intelligence startup, has launched the AI Materials Foundry and secured $450 million（約720億円） in Series B funding backed by investors including Jeff Bezos' investment fund. The consortium brings together 48 technology, industrial and research organizations—including Nvidia, Meta Platforms, and Hyundai Motor Group—to accelerate discovery of advanced materials for semiconductor manufacturing and other industrial applications.

The initiative combines computing infrastructure with scientific expertise to develop software aimed at reducing the time and cost required to identify next-generation materials used in chips. CuspAI will use part of the funding to expand laboratory operations with foundry partners located in Cambridge, Singapore, and the San Francisco Bay area. The company's strategic focus on semiconductors represents a shift from its original mission; CuspAI was initially concentrated on materials for carbon capture and water purification but pivoted toward the semiconductor industry as demand from the AI chip supply chain expanded.

Nvidia stated it expects to collaborate with CuspAI and other alliance members on materials research projects. Meta joins the coalition as part of its broader efforts to support AI infrastructure development. The consortium's formation reflects growing recognition across the industry that advanced materials have become a critical enabler for semiconductor performance and that expanded opportunities exist across the semiconductor sector to accelerate discovery and deployment of these materials.

CuspAI's shift from carbon capture and water purification to semiconductor materials reflects a strategic pivot driven by rising demand in the AI chip supply chain. The consortium model—bringing together 48 organizations spanning technology giants (Nvidia, Meta), automotive (Hyundai Motor Group), and research institutions—signals that material science has become a critical bottleneck in semiconductor advancement. By pooling computing infrastructure and scientific expertise across Cambridge, Singapore, and the San Francisco Bay area, the initiative aims to compress timelines and reduce costs in a discovery process typically considered slow and expensive. Nvidia's commitment to collaborate on research and Meta's participation as part of its broader AI infrastructure strategy underscores that chipmakers and AI providers view advanced materials as essential to supporting the next generation of semiconductor performance.