Drop

Einstein's own version of the equivalence principle is a thought experiment: shut in a windowless box, you can't tell whether you're sitting on a launch pad under gravity or being accelerated through empty space at the same rate — free fall and weightlessness look identical from the inside. It's held up against every classical test thrown at it for over a century. What had never been done, until this month, was watching it hold for a single quantum object — not a thing falling, but a wave.

An atom in quantum superposition doesn't take one path; the same atom takes two at once, and the two copies can be steered apart and reunited. A team led by Ron Folman at Ben-Gurion University of the Negev, with the University of Ulm, the University of Oxford — including Sir Roger Penrose and Vlatko Vedral — and others, built exactly that out of ultracold rubidium atoms on a chip: split each atom in two, hold one half level against gravity with a magnetic field, let the other half go and fall freely, then bring the two halves back together. Recombined, the two copies interfere, and the interference encodes a phase — a number gravity should have imprinted on the falling half and not the held one, if the equivalence principle reaches all the way down to a single atom's own quantum wave. It does, and this “Quantum Galileo Interferometer” is the first apparatus to measure that phase directly, published 2026-09-02 in Science Advances.

The button only ever changes the two atom markers and the fringes at the bottom — the chip and the diamond of dashed arms are fixed, the same schematic either way.

What this does and doesn't show. It confirms the equivalence principle keeps holding once you're asking it about a single quantum wave rather than a falling ball, for atoms this light, over this short a fall. It does not show gravity itself is quantized, and it doesn't unify quantum mechanics with general relativity — as Vedral put it, “we have no consistent theory telling us why quantum physics should fail,” and this experiment narrows, rather than closes, that gap. It also isn't a test of Penrose's own older proposal that quantum superposition itself should break down under gravity for large enough, long-lived objects — light rubidium atoms in a brief superposition never reach the mass or timescale that idea needs. The same team is already chasing that next, at Ben-Gurion, with something far heavier: nanodiamonds.

Honest gap: the diagram above is a schematic, not a floor plan. The real two paths are separated by height, not by left and right — one atom held at a fixed height by the chip's field, the other released to fall straight down — drawn here as a diamond only so both arms fit in one small picture without crossing. The fringe pattern at the bottom stands for “a phase difference was measured,” not the interferometer's own actual readout.

Sources: Dobkowski, O., Folman, R., et al., “Observation of the quantum phase of free fall and the consistency with the equivalence principle,” Science Advances, 2026-09-02 (DOI 10.1126/sciadv.aec8045); reporting via the University of Oxford Department of Physics and phys.org.