In a groundbreaking development, researchers at Emory University have used artificial intelligence to uncover a previously unknown fundamental law of physics, signaling a major advancement in the way scientific knowledge is discovered and validated. The team applied machine learning to analyze highly complex experimental data from a form of ionized gas known as dusty plasma, ultimately revealing non-reciprocal forces between particles that were not explained by existing theories.
This novel approach, detailed in the Proceedings of the National Academy of Sciences, marks a critical evolution in scientific methodology. Rather than simply analyzing known data or verifying hypotheses, the AI system generated new, testable laws of nature based on real-world experimental results. The machine learning model was trained to observe how particles in dusty plasma systems interacted and adjusted its predictions to achieve over 99 percent accuracy in modeling their behaviors. The findings challenge longstanding assumptions in plasma physics and offer more precise descriptions of how particles exert forces on one another under extreme conditions.
Lead researcher Justin Burton, professor of physics at Emory, emphasized the significance of the discovery. He noted that unlike traditional AI “black box” models, the system used here produced transparent and interpretable equations that physicists can understand and apply. His colleague Ilya Nemenman, a theoretical physicist on the project, added that the model’s ability to correct the flaws in previous frameworks exemplifies how artificial intelligence can serve not just as a tool for analysis, but as a true partner in theoretical innovation.
Beyond its impact on plasma physics, the discovery has broader implications for other complex systems in nature, particularly in the study of neutrinos—mysterious particles that pass through matter with minimal interaction. By refining our understanding of many-body systems like dusty plasma, the AI-assisted method could help scientists develop more accurate theoretical models of neutrinos and design better experiments to detect and study them. These developments are especially relevant as the scientific community prepares to launch next-generation particle physics experiments over the coming years.
Experts across the scientific community see the Emory team’s work as a turning point. It demonstrates that AI can do more than recognize patterns—it can generate scientific knowledge that expands the boundaries of human understanding. This approach may redefine how new laws of nature are discovered, with human ingenuity and artificial intelligence working side by side to uncover the universe’s deepest secrets.
