Discovered Materials, a recent San Francisco startup utilizing artificial intelligence to identify novel compounds for semiconductor applications, has announced a substantial incentive program. The company is offering a total prize pool of $1 million to AI systems capable of resolving five critical, open-ended challenges in the field of materials science.
While cutting-edge AI models have successfully addressed complex mathematical puzzles and made strides in biological research, they have yet to fully master the intricate physics inherent in advanced materials.
"It is more difficult to articulate problems clearly in materials science because it is not like mathematics, where there is a precise definition and a definitive yes or no," explained Advaith Sridhar, a co-founder of Discovered Materials, which is part of Y Combinator’s Class of 2026. "We entered this discussion [wondering], what are the grand challenges in this field? How do we know AI is truly making progress?"
Although materials science lacks a direct equivalent to the Millennium Prize Problems, Discovered Materials has engaged an advisory board of academics from prestigious global universities to define these five hurdles. The organization is distributing a combined $1 million in awards to anyone who can provide solutions.
"Across materials science, there are phenomena we observe but cannot explain," stated Felipe Jornada, a Stanford professor involved in developing the challenges. "Explaining these phenomena would enhance our understanding of physics, and the results would directly influence the development of electronics, energy storage, and the manufacturing of future materials."
Prior to the announcement, Discovered Materials utilized frontier laboratory models to attempt autonomous solutions, providing these AI agents with internet access and resources over several days. The models progressed slowly on each issue. Nevertheless, it is plausible that entities like OpenAI or Anthropic might achieve a breakthrough using significantly more computational power or a newer iteration of their technology.
"It is not particularly a concern to us whether we find it, or someone else finds it, or a professor finds it, or the frontier labs find it," Sridhar remarked. "The motivation behind these challenges is to establish benchmarks. If someone solves them, it deepens our collective understanding of materials and the behaviors of the world around us."
Monitoring the efforts to solve these problems will indicate "whether the models are climbing in this domain," he added.
This challenge highlights the evolving nature of scientific discovery, as AI demonstrates the ability to uncover insights and methodologies that human researchers never anticipated. With solutions appearing increasingly accessible, the questions themselves become more vital.
"This represents a novel approach to scientific inquiry, where we establish verification criteria and challenges first, and then invite individuals to submit ideas that elucidate the problems," Sridhar said. "Determining the appropriate questions to ask is non-trivial and highly challenging in this domain, which is why the professors have been instrumental in helping us frame the correct inquiries."
The primary mission of Discovered Materials is the creation of new substances for semiconductor use. The startup collaborates with major chip manufacturers on research initiatives while also funding its own scientific endeavors. When the company disclosed its seed funding round last August, Sridhar indicated a goal to generate a commercially viable material within one year.
"Those timelines have accelerated slightly," he notes today. "I am more confident now that we will be able to identify industrially useful materials within a timeframe shorter than a year."
The Editorial Staff at AIChief is a team of professional content writers with extensive experience in AI and marketing. Founded in 2025, AIChief has quickly grown into the largest free AI resource hub in the industry.