A San Francisco-based startup, Discovered Materials, co-founded by IIT Madras alumni Advaith Sridhar and Akash Ramdas, is offering a $1 million prize to individuals, teams, or organizations capable of solving five persistent challenges in materials science. The initiative aims to accelerate scientific discovery and establish new benchmarks for progress in the field, including the capabilities of artificial intelligence.
The company, which recently secured $9 million in seed funding and participated in Y Combinator's spring 2026 batch, collaborated with leading academic institutions to identify these complex problems. Despite years of intensive research, these fundamental questions have remained unanswered.
The Five Grand Challenges in Materials Science
The challenges span critical areas within materials science, each holding significant implications for future technological advancements:
- Cuprate Superconductivity: Understanding the mechanisms behind high-temperature superconductivity in cuprates, which offers the largest prize of $400,000.
- Semiconductor Interface Properties: Accurately predicting the electrical characteristics at the interfaces of different semiconductor materials.
- Quantum Chemistry Calculations: Developing faster and more precise methods for complex quantum chemistry computations.
- Light-Material Interaction: Explaining how light fundamentally alters the behavior and properties of various materials.
- Heat Transfer Across Interfaces: Predicting the movement of heat between solid materials at their interfaces.
Solving these problems could lead to breakthroughs in developing more efficient electronic devices, revolutionizing semiconductor manufacturing processes, and advancing critical energy-related technologies.
AI Models Stumble Where Humans Might Prevail
Before opening the competition to the public, Discovered Materials subjected advanced AI models to these same problems. The AI agents, operating autonomously with access to vast internet resources, made minimal progress even after several days of independent work. This outcome underscores the profound difficulty of these scientific puzzles and highlights the current limitations of AI in truly novel discovery.
As Stanford Professor Felipe Jornada, who assisted in formulating the challenges, noted, "Across materials science, there are phenomena we observe but cannot explain." For the founders of Discovered Materials, the competition is not just about finding solutions but also about creating a measurable framework to assess genuine progress in scientific inquiry, both by human researchers and artificial intelligence.