Quantum computer completes verified task beyond practical reach of classical simulations

Phys.org ^ | 08/01/2026 | University of Chicago

[BenLurkin]

IBM and researchers from the University of Chicago announced a demonstration in quantum computing that meets the fundamental criteria for "quantum advantage"—the point where quantum computers can be confirmed to have outperformed classical computers on trusted computations.

The collaboration said its system had performed computations beyond the reach of leading classical simulation methods while providing confidence that the computation returned accurate results.

In their new paper, the researchers showed that these two goals could be achieved through a novel construction of encoded quantum circuits—one of the largest demonstrations of logical quantum computing to date. The paper is published on the arXiv preprint server.

Building trust into quantum results For years, researchers have used a benchmark known as random circuit sampling, or RCS, to test whether quantum computers could outperform classical systems.

In simple terms, RCS asks a quantum computer to generate patterns so complex that a classical computer cannot efficiently reproduce them.

The challenge has been verification: As the problem becomes harder, it becomes increasingly difficult, then infeasible, to prove the quantum computer's answer is correct without making strong assumptions about the inner workings of the quantum computer.

In their experiment, researchers addressed this obstacle with a structured alternative to RCS. The team was able to prove that this alternative retains the same hardness criteria as RCS, but crucially, the new structure can be used to detect errors during the computation

[Rockingham]

Quantum computing will be uniquely powerful for a relatively limited set of important computational tasks that are unsuited to the constraints of conventional computing. Material science will see major gains.

[Honorary Serb]

New! (Pronounced “Nyew”.)

[Scrambler Bob]

Calculate infinity to the infinity to the infinity to the infinity to the infinity to get warmed up.

[Reily]

Easy! Its infinity!

[Right_Wing_Madman]

The biggest problem is decoherence where even a stray molecule or photon will drop the superpositioning and entanglement.

In order to eliminate stray particles, a temperature of absolute zero is needed, and infinite energy is required to reach a temperature of true zero. Quantum computing, by design, will never be 100 percent accurate, but can it be 99.999999% accurate?

[captmar-vell]

That has been the most well known, and I think IBM /Google is still leading with it.

But there are other approaches not requiring the same amount of extreme hardware, none are clear leaders yet, maybe Ionq with their trapped-ion research, then there is qubit and last quantum annealing technology.

[Some Fat Guy in L.A.]

The answer was 42.