Discovered Materials, a San Francisco-based startup leveraging AI to discover new materials for semiconductor manufacturing, announced a $1 million prize pool targeting five unsolved problems in materials science. The company recruited academic advisors from leading universities to define critical challenges including cuprate superconductivity theory, silicon interface prediction methods, thermal property mapping, photoinduced behavior explanation, and improvements to Density Functional Theory. After testing frontier AI models on these problems without significant progress, the startup aims to establish measurable benchmarks for evaluating AI advancement in the materials science domain. The initiative reflects a shift in scientific methodology where verification frameworks and problem definition precede solution discovery.