A group of AI agents can now watch raw data from a physical system and write down the exact math equations that govern it.
This multi-agent system acts as an automated scientist that simplifies massive amounts of information into human-readable laws. Standard AI usually functions as a black box with millions of parameters that nobody can explain. This new approach cuts those millions of variables down to between 5 and 40 interpretable pieces. It successfully extracted the governing physics from complex datasets that would normally require years of human study. This technology turns opaque machine learning into a tool that explains why the universe works the way it does.
Machine Collective Intelligence for Explainable Scientific Discovery
arXiv · 2604.27297
Deriving governing equations from empirical observations is a longstanding challenge in science. Although artificial intelligence (AI) has demonstrated substantial capabilities in function approximation, the discovery of explainable and extrapolatable equations remains a fundamental limitation of modern AI, posing a central bottleneck for AI-driven scientific discovery. Here, we present machine collective intelligence, a unified paradigm that integrates two fundamental yet distinct traditions in