Chemistry-focused CuspAI raises $100M to accelerate AI-designed materials discovery

CuspAI, an artificial intelligence startup developing models for chemistry discovery, has raised $100 million in Series A funding led by New Enterprise Associates (NEA), a prominent venture capital firm, and Temasek, Singapore’s sovereign wealth fund. The round, which values the company at $520 million, positions CuspAI to accelerate the development of AI-designed materials ranging from carbon capture substances to advanced semiconductors, targeting markets that could reshape manufacturing and sustainability efforts.

What you should know: CuspAI has assembled an impressive roster of AI luminaries and industry veterans to guide its mission of using AI to discover new materials.

• The company’s advisory board includes AI “godfathers” Geoffrey Hinton and Yann LeCun, both Turing Award winners, alongside materials science expert Kristin Persson from UC Berkeley.
• New additions include former BP CEO Lord John Browne and Martin van den Brink, former president and CTO of ASML, a company that makes the lithography technology for producing advanced semiconductors.
• The startup has hired talent from leading AI labs including Google DeepMind and Meta, building a team of about 30 employees.

The big picture: CuspAI is building AI “foundation models” for chemistry that function like search engines for discovering new materials with specific properties.

• The company describes its technology as “a search engine for the material world,” making predictions about chemistry similar to how large language models predict language.
• Working with Meta and Georgia Tech, CuspAI helped create OpenDAC, the world’s largest dataset on materials for carbon dioxide capture, containing over 100 million datapoints.
• CEO Chad Edwards says the company can now go from customer specifications to a completely novel material within six months, a process that typically takes a decade.

Key partnerships: The funding round brought strategic investors who are also becoming commercial partners.

• Hyundai Motor Group is partnering with CuspAI on “sustainable energy applications,” potentially involving battery materials or hydrogen fuel cell improvements.
• Water company Kimera has signed an agreement to find materials that could filter PFAS “forever chemicals” from water supplies.
• Edwards indicated the company is exploring partnerships in the semiconductor industry as well.

Competitive landscape: The materials science AI sector is experiencing rapid growth and intense competition for funding.

• When CuspAI announced its $30 million seed round in June 2024, it was the largest for a materials science AI startup at the time.
• Since then, Flagship Pioneering spun out Lila Sciences with a $200 million seed round, while Periodic Labs raised $200 million from Andreessen Horowitz at a $1 billion valuation.
• Edwards emphasized that CuspAI’s “synthesis-aware generative AI models” are designed to produce materials that can actually be manufactured, addressing a common problem with AI-generated material recipes.

What they’re saying: Investors highlighted the company’s technical advantages and execution speed.

• “At Microsoft, we used to say what really matters in breakthroughs in AI is scale and speed. And I think this team has both,” said Lila Tretikov, NEA partner and former Microsoft deputy CTO who led the investment.
• “We believe novel materials offer an extraordinary opportunity to accelerate progress towards a better, cleaner, and more efficient future,” said Keith Noh, vice president of Hyundai Motor Group’s ZERO1NE division.

What’s next: CuspAI plans to use the funding to expand its team and open new offices, with Asia as a particular focus.

• The company aims to reduce its material development timeline from six months to just one to two months within the next two years.
• Edwards said CuspAI is pursuing a dual strategy of creating custom materials for large partners while developing materials for its own projects, though the current emphasis remains on partner collaborations.