Gravity Still Sucks — But Researchers Say Quantum Interference Could Make it Push

Insider Brief

• Physicists have proposed a tabletop experiment in which quantum interference between two gravitational pulls could produce an effective repulsive momentum shift, offering a potential test of whether gravity is fundamentally quantum.
• The scheme requires placing only one mass in a spatial superposition and relies on weak-value amplification and post-selection to make the otherwise tiny gravitational effect measurable.
• Although technically challenging and still theoretical, the proposal outlines parameter ranges within reach of advanced nanodiamond, ultracold atom and interferometry platforms now under development.
• Photo by meriç tuna on Unsplash

Gravity may be able to push instead of pull — at least in a exquisitely and very carefully engineered quantum experiment — according to a new theoretical proposal that also outlines a simpler path to testing whether gravity itself is fundamentally quantum.

Before anyone starts investing in anti-gravity machines or an anti-gravity machine startup, the proposal does not suggest that gravity itself becomes repulsive; rather, quantum interference — which we’ll touch on soon — between two possible gravitational pulls can produce an average momentum shift in the opposite direction.

In the paper posted on the pre-print server arXiv, physicists Pablo L. Saldanha of the Federal University of Minas Gerais and Chiara Marletto and Vlatko Vedral of the University of Oxford describe a tabletop experiment in which a single mass placed in a quantum superposition could cause a second particle to experience an effective gravitational repulsion. Such a result would have no classical explanation and would serve as evidence that gravity possesses quantum properties, according to the researchers.

The proposal builds on a growing research program aimed at probing gravity at small scales using quantum technologies rather than high-energy particle colliders or cosmological observations. For decades, physicists have struggled to reconcile quantum theory — which governs atoms and subatomic particles — with general relativity, which describes gravity as the curvature of spacetime. While the other fundamental forces have been successfully incorporated into quantum theory, gravity remains the outlier.

The new work refines earlier ideas that sought to detect “gravitationally induced entanglement,” a phenomenon in which two tiny masses placed in quantum superpositions of position become entangled solely through their mutual gravitational attraction. Entanglement is a distinctly quantum effect in which particles become linked so that measuring one instantaneously influences the other. Because classical forces cannot create entanglement, observing it would strongly suggest that gravity itself must carry quantum degrees of freedom.

Previous proposals required placing two relatively massive objects into spatial superposition — a technically daunting task. The new scheme reduces that burden by requiring only one mass to be placed in a superposition. The second object, known as the probe, can remain in an ordinary quantum state without being split between two locations.

Quantum Interference

The central idea relies on quantum interference, a well-established phenomenon in which different possible outcomes combine in ways that can enhance or cancel each other. In the proposed experiment, a “source” mass is placed in a superposition of two positions inside a Mach-Zehnder interferometer, a device commonly used in optics and atomic physics to split and recombine wave-like particles. While in superposition, the source exerts a gravitational pull on a nearby probe particle.

If gravity obeys the quantum superposition principle, the probe should experience a superposition of two gravitational attractions — one from each possible location of the source mass. Under certain conditions, and with a carefully chosen postselection of the source’s final quantum state, the resulting interference could cause the probe to receive a momentum kick opposite to the direction of the gravitational pull.

In classical physics, gravity is always attractive between two positive masses. There is no known mechanism by which gravity could produce a repulsive force between ordinary matter. In the proposed quantum scenario, however, the average momentum transfer to the probe particle could be negative — meaning it moves as if pushed away. The effect does not imply that gravity becomes fundamentally repulsive. Instead, it reflects interference between two quantum alternatives of the gravitational field.

The researchers show mathematically that this anomalous momentum transfer arises from entanglement between the source and probe particles mediated by gravity. In their formal treatment, they analyze the interaction using both the standard Schrödinger picture and the Heisenberg picture of quantum mechanics. In both descriptions, the outcome depends on what is known as a weak value — a quantity that can exceed the range of ordinary measurement results when a quantum system is preselected in one state and postselected in another nearly orthogonal state.

Weak values have been studied for decades and have been demonstrated in optical experiments. In this proposal, they allow the small gravitational effect to be amplified, potentially by orders of magnitude. The amplification, however, comes at a cost: the probability of successfully postselecting the desired state becomes correspondingly small.

The paper includes preliminary feasibility estimates. Earlier gravitational entanglement proposals envisioned nanodiamonds with embedded nitrogen-vacancy centers, weighing around 10^-14 kilograms — or, one hundred-trillionth of a kilogram, placed into superposition across distances of tens to hundreds of micrometers. Using similar parameters, the new scheme finds that observing a measurable effect may require either increasing the mass of the source particle or adjusting other experimental variables.

One scenario described in the study suggests that with a source mass of roughly 10^-14 kilograms and a probe mass of 10⁻²⁰ kilograms — or, one hundred-quintillionth of a kilograms, combined with micrometer-scale separations and interaction times of fractions of a second, the anomalous momentum shift could reach about 0.2 percent of the probe’s intrinsic momentum uncertainty. According to the team’s estimates, a shift on the order of 0.1 percent should be measurable using established techniques for detecting momentum distributions in ultracold atoms or Bose-Einstein condensates.

Limitations And Future Work

The proposal also highlights practical challenges and room for future work. For example, because the gravitational force between such small objects is extraordinarily weak, competing forces — such as electromagnetic interactions or forces arising from quantum fluctuations between closely spaced objects (called Casimir–Polder forces) — could mask the signal. Distinguishing a genuine gravitational effect from these backgrounds would require careful experimental design and shielding.

Another limitation is the reliance on weak-value amplification. While it enhances the observable signal, it reduces the number of successful measurement runs, since most attempts would not meet the required post-selection criteria. This tradeoff is familiar from other weak measurement experiments but adds complexity when dealing with already tiny gravitational effects.

Vedral addresses post-selection in a Substack post: “‘Oh, well, anything can happen if we post-select,’ I hear you say. Yes and no. Indeed, throwing away ‘bad’ outcomes (like the ones where the particles attract) leads us to observe what we want (i.e., repulsion), however, classically, this is not possible, no matter how much we post-select. Our experiment, if confirmed experimentally, would therefore show that gravity can act from two different places at the same time. In other words, gravity is quantum.”

Despite these hurdles, the researchers report that their scheme offers conceptual and practical advantages over earlier proposals. By requiring only one mass in superposition and eliminating the need to directly measure entanglement correlations between two large objects, the setup simplifies some of the most demanding aspects of previous designs.

Practical and Broader Implications?

For the quantum technology sector, the proposal might be sitting at a strategic point — it’s at the edge of current capabilities but not necessarily far beyond them. Experiments involving nanodiamonds with nitrogen-vacancy centers, ultracold atoms and precision interferometry are already under development for sensing and quantum information applications. The new scheme repurposes many of those same tools, but pushes them into a regime where gravitational effects must be isolated from electromagnetic noise at unprecedented sensitivity. While not immediately commercial, such experiments would extend the technical frontier of quantum control and metrology, areas that underpin much of the industry’s long-term roadmap.

The broader implication is that quantum gravity may not require Planck-scale energies to test. For much of the past century, probing quantum aspects of gravity was thought to be possible only at extreme energy scales far beyond laboratory reach. The recent wave of proposals, including this one, suggests that carefully controlled quantum systems at the micrometer and millisecond scale may suffice to reveal whether gravity can carry quantum information.

If an experiment were to observe the predicted effective repulsion, it would not provide a complete theory of quantum gravity. It would, however, offer strong evidence that the gravitational field cannot be purely classical. That conclusion would support approaches in which gravity itself is described by quantum states.

Future work will likely focus on refining parameter estimates, analyzing environmental noise and decoherence, and designing experimental platforms capable of isolating gravitational interactions at unprecedented sensitivity. Advances in quantum control of nanomechanical systems, ultracold atoms and solid-state defects may bring such tests within reach.

The paper is just about four pages long, but it’s incredibly deep and technical and much of it is beyond the scope of this article. For a deeper, more technical dive, please review the paper on arXiv. We should also note that arXiv is a pre-print server, which allows researchers to receive quick feedback on their work. However, it is not — nor is this article, itself — official peer-review publications. Peer-review is an important step in the scientific process to verify results.