Microsoft, Atom Computing, and EeroQ push quantum forward

Three separate efforts show quantum computing inching closer to reliability: Microsoft reports logical qubits kept stable through error correction for up to 90 rounds, Atom Computing contributes progress that narrows the gap to practical error correction, and

For quantum computing, the hardest part isn’t getting qubits to behave like qubits. It’s keeping them behaving that way long enough for a calculation to mean anything.

One of the latest steps comes from Microsoft’s work on error correction. The progress doesn’t magically erase the problem—eventually. an error arrives that can’t be recovered from because too many individual atoms change state at once. Still, performing normal error correction can keep some logical qubits stable for up to 90 rounds. It’s not enough for sophisticated computation yet. but it’s “a lot closer” than the company was before developing the technique that made those longer stretches possible.

That tension—between what quantum can do today and what it must do to be useful—runs through the rest of the field.

Atom Computing’s update sits in the same reality check. Its work reflects the challenge that quantum systems don’t fail politely. When errors pile up faster than the correction can untangle them, the logical qubit state can’t be restored. The progress is about shrinking that gap. even if stability still falls short of the threshold needed for large-scale. error-corrected calculations.

EeroQ is taking a different route to the same destination: usable qubits that can be controlled reliably.

The startup’s approach uses the spin of electrons as qubits. The attraction is practical: it’s relatively easy to fabricate chips that can manipulate electrons held in quantum dots. EeroQ builds its chips with lots of tiny pools designed to hold a drop of liquid helium. When an electron is placed on that helium surface. it can’t easily go anywhere—helium “hates” to carry an extra electron. so the electron effectively floats on the surface.

The catch is that this physics was already known long before EeroQ started. The unresolved question was how to interact with the electron in ways that could become something programmable.

Recently, the company released a manuscript describing a new version of its chip. This one places a small resonator next to the helium-filled pool. In experiments, the resonator can couple with the electron’s movement. That movement is kept from slamming into the walls of the pool by an electrical field. Because the electron’s motional states are quantized. the resonator can adopt one or two states during the experimental procedure—states that are described as potential building blocks for a qubit.

Again, none of this turns into a working quantum computer overnight. But it does show the kind of steady engineering that the field has been missing: building mechanisms that connect fragile quantum behavior to controllable hardware elements.

The three updates read like different chapters of the same story—errors show up. systems need shielding and correction. and new qubit designs still have to prove they can be driven into states that matter. Microsoft’s 90-round window doesn’t solve quantum computing. but it does extend the time in which quantum information can be held onto. Atom Computing’s focus on the limits of recovery points to where robustness is still missing. And EeroQ’s resonator-coupled helium electron experiments show how the path toward a qubit can be carved out of known physics.

In quantum computing, the breakthrough often doesn’t arrive as a single leap. It arrives as a series of small, stubborn improvements—until the moments of stability finally become long enough to carry a real calculation.

quantum computing error correction logical qubits Microsoft Atom Computing EeroQ helium-based qubits resonator electron spin quantum dots