Bezos-backed CuspAI Raises $450 Million to Use AI for Discovering New Materials for Chipmakers

TLDR

• CuspAI raised $450 million in a Series B round, valuing the startup at $2.6 billion
• Investors include the UK government, Jeff Bezos’ fund, AMD Ventures, and Kleiner Perkins
• The company launched AI Materials Foundry, a coalition of 48+ firms including Nvidia and Meta
• CuspAI’s AI platform MIRA aims to find new materials to reduce reliance on rare metals like ruthenium and iridium in chipmaking
• So far, none of CuspAI’s projects have produced a candidate molecule ready to leave the lab

---

Britain’s CuspAI has closed a $450 million Series B funding round, pushing its valuation to $2.6 billion. That’s up from $520 million just last September.

Introducing CuspAI’s AI Materials Foundry.

We’re launching a global network to fundamentally accelerate how we discover and design the physical building blocks of our world. pic.twitter.com/2pcj9J6JWD

— CuspAI (@cusp_ai) July 20, 2026

The round was led by venture firms Kleiner Perkins and NEA. Jeff Bezos’ investment fund, Bezos Expeditions, participated alongside the UK’s Sovereign AI Venture Fund, AMD Ventures, Glade Brook Capital Partners, Lux Capital, and the Netherlands’ Invest-NL.

The Cambridge-based startup is betting that artificial intelligence can speed up the discovery of new materials for chipmakers and other industries.

AI Materials Foundry Launches With 48+ Members

On Monday, CuspAI also launched the AI Materials Foundry. The coalition includes more than 48 companies and research institutions, including Nvidia, Meta, and Hyundai Motor Group.

The goal is to pool computing and scientific resources to build software that helps researchers develop new materials faster and at lower cost than current methods.

CuspAI’s platform, called MIRA, runs full discovery cycles — from generative materials design through simulation and experimental validation.

Targeting Rare Metals in Chipmaking

One of CuspAI’s main targets is reducing or eliminating the use of rare metals like ruthenium and iridium in chipmaking. These materials carry supply chain risks that chipmakers are eager to work around.

CEO Chad Edwards said chipmakers and semiconductor suppliers are “frantically searching for new materials.” He added the company has been “literally pulled by all four limbs” by demand from the chip supply chain.

CuspAI originally focused on carbon capture and water purification before pivoting toward chip materials over the past year.

Nvidia’s Geetika Gupta said the chipmaker sees demand for new materials across energy storage and data center cooling. She said Nvidia will likely work through the Foundry in three-way collaborations with CuspAI and a third partner.

New hires include John Giannandrea, formerly of Apple and Google, who is helping set up CuspAI’s California operations. AMD board member and semiconductor veteran Abhi Talwalkar has joined its advisory board alongside AI pioneers Yann LeCun and Geoffrey Hinton.

CuspAI plans to use most of the new funding to support labs in Cambridge, Singapore, and the San Francisco Bay Area.

Early Results Mixed

Despite the high-profile backing, none of CuspAI’s projects have yet produced a candidate molecule ready to move beyond the lab.

An early carbon capture project with Meta researchers identified 10 promising structures from a library of 300 trillion. Fewer than 10 were synthesized, and none outperformed what was already commercially available.

The startup is now using the same library to find materials that can remove toxic “forever chemicals” from water. Finnish chemicals firm Kemira will test 20 of these new structures this year.

Co-founder Max Welling said limited data, testing facilities, and computing power have held the field back — problems CuspAI hopes to solve with its foundry network.