Discovered Materials raises $9M to find cooler chip materials with AI

In short: Discovered Materials raised $9 million to use AI to search for new materials that could help computer chips run cooler and waste less electricity.

What happened

Discovered Materials, a startup, said it closed a $9 million seed funding round led by Lightspeed India Partners. It recently came out of Y Combinator, a well-known startup program. Peak XV Partners and angel investors including Paul Graham, Gokul Rajaram, and Thariq Shihipar also invested.

The company is trying to tackle a growing problem in data centers, which are the buildings full of computers that run online services. Chips running AI tasks can get very hot, and that heat means more electricity is needed for cooling.

Discovered Materials says it uses “AI agents,” which are software workers that can run tasks on their own (like having lots of tireless lab assistants). These agents propose possible new materials, then the company runs computer simulations to check whether those materials might have the right properties. The founders say this lets them test thousands of ideas a day, compared with a few dozen guesses a day during a traditional PhD research process.

The startup also released examples of hundreds of new materials and a tool it calls “Material Discovery Bench,” meant to track how well advanced AI models do at this kind of search.

Why it matters

If chips can run cooler, data centers may need less power and less cooling equipment, which can reduce costs and energy use. Still, investors and the founders noted a key limitation: even if software suggests a promising material, it still has to be made in a real lab and then manufactured into chips at scale, and that step is slow.

Source: TechCrunch AI