Gravity from entropy: how a London mathematician is reframing the universe's first breath

On 17 July 2026 a team led by Queen Mary University of London mathematician Professor Ginestra Bianconi put forward a sweeping mathematical argument: that gravity, the weakest of the four fundamental forces, can be derived from the second law of thermodynamics, the same rule of thumb that dictates why ice melts and engines cannot run forever. The proposal, covered by PHYS on the same day, recasts one of physics' deepest puzzles as a question about information, geometry, and the statistical behaviour of large numbers of quantum components.

The work lands inside a long-running effort to reconcile general relativity, Einstein's geometric theory of gravity, with quantum mechanics, the rules governing the subatomic world. Those two pillars of modern physics have resisted unification for a century. Bianconi's move is not to write a new theory of everything from scratch, but to show that gravity's familiar field equations can fall out of a purely entropic accounting, the same accounting that governs how heat disperses and how disordered a system can become.

What the new framework actually claims

The paper treats spacetime as a network of quantum components whose configurations can be counted. The second law says entropy, a measure of how many microstates a system can occupy, tends to grow over time. By demanding that entropy increase on a curved geometry, the mathematics reproduces equations of the same family as Einstein's. In other words, gravity is what the universe looks like when you demand that disorder increase consistently across a vast, structured arena.

The approach borrows from the so-called entropic-gravity programme that has circulated for years, in which gravitational attraction is reframed as a statistical tendency rather than a fundamental pull. The contribution here is technical: a rigorous derivation that does not assume classical spacetime to begin with, and that places the argument on the same graph-theoretic footing now used to model complex networks in biology and computer science.

Why this is harder than it sounds

Einstein's equations are notoriously unforgiving. They describe the bending of light around stars, the expansion of the universe, and the ripples in spacetime known as gravitational waves, phenomena confirmed to extraordinary precision. Any alternative derivation has to reproduce that record while sitting comfortably alongside quantum theory.

The entropic route has struggled, until now, to generate the dynamics: how gravity changes over time, not just how matter arranges itself in space. The new paper claims to bridge that gap by tying the network's evolution to entropy maximisation at every step. The mathematics is intricate enough that independent groups will need months to check it. Reproducibility in theoretical physics is a slow process: derivations are re-derived, boundary cases tested, and disagreements aired in journals before a community treats a result as canonical.

What changes if the result holds

If subsequent work confirms the derivation, the implication is not that Einstein was wrong, but that his equations can be read as a theorem about information flow rather than as a fundamental law of nature in their own right. That would be a meaningful philosophical shift, and a practical one: it would open the door to a fresh class of calculations about the very early universe, where quantum effects dominate and gravity is currently hardest to apply.

A second-order consequence is conceptual. Cosmology's standard narrative holds that the universe emerged from a hot, dense state and has been expanding and cooling ever since. That story already leans heavily on thermodynamics. If gravity itself is entropic, the second law is no longer just a description of what matter does in space; it is part of the instruction manual for how space itself behaves. The bridge between the very large and the very small becomes a single document rather than two that have to be stapled together.

What remains uncertain

The result is, at this stage, a theoretical proposal supported by a mathematical derivation. It has not yet been tested against astronomical data, nor has it generated new predictions that distinguish it from standard general relativity. The community's standing reaction to claims of this kind is to ask three questions: does the derivation cover all the cases general relativity does, does it make a testable new prediction, and does it survive peer review when independent groups re-do the algebra.

None of those answers are in yet. The framework is, however, unusually clean: it uses a language, graph theory and entropy maximisation, that physicists already trust in adjacent fields. That alone makes it more than a curiosity. Whether it becomes a foundation or remains an elegant dead end will be settled by work over the next several years, not by press release.

How Monexus framed this: the wire coverage focused on the headline claim, gravity from entropy. The desk note here is that the same claim has been made in various forms for more than a decade, and the value of this paper sits in the rigour of the derivation rather than in the novelty of the slogan.