A YC startup is putting $1M behind five unsolved materials science problems for AI
Reported by TechCrunch1 source ↓
Discovered Materials is offering a total prize pool of $1 million for solutions to five open materials science problems, with the aim of tracking AI progress in the field.
Key points
- The five challenges were set out with help from academic advisers
- AI agents tested by the company spent several days on each problem without making much headway
- Discovered Materials develops materials for semiconductors and works with chip companies on research
Sources
Details
The San Francisco startup, part of Y Combinator’s spring 2026 batch, assembled an academic advisory group to define five open problems. The article does not specify how the $1 million prize pool is divided among them.
Before announcing the challenges, the company had AI agents with internet access and other resources work on each problem for several days. TechCrunch reports they made little headway; Discovered Materials says the challenges are intended as goalposts for measuring progress in AI-driven materials science.
The company also funds its own scientific research and partners with chip companies. Co-founder Advaith Sridhar told TechCrunch he was more confident the company could find industrially useful materials in less than a year, but the article does not report that it has done so.
Written from coverage as of
Images
- TechCrunch