D-Wave’s Quantum Supremacy Claims Challenged by Researchers

VANCOUVER, Canada — Quantum computing firm D-Wave has asserted that its Advantage quantum computers can solve problems that are virtually impossible for traditional computers. However, recent research from two independent groups has raised doubts about these claims, suggesting that the problems can be tackled using classical algorithms.

D-Wave’s assertions were detailed in a recently published preprint that described successes in calculating transverse field Ising model problems, which are quantum representations of phenomena such as phase transitions, including changes from liquid to gas. The company claimed that these computations would take traditional computers impractically long to solve. This research subsequently passed peer review and was published on March 12, 2025.

Researchers at New York University, led by Dries Sels, challenged D-Wave’s claims utilizing a mathematical framework known as tensor networks. This approach optimizes the data required for simulations, dramatically reducing the computational resources necessary. In response, D-Wave stated that their results remain unchanged, as the NYU team did not replicate all parameters or observable outcomes from their experiments.

“They didn’t do all the problems that we did, they didn’t do all the sizes we did, they didn’t do all the observables we did, and they didn’t do all the simulation tests we did,” said David King, a representative from D-Wave. King indicated that he conducted trials with calculations involving up to 3,200 qubits, far surpassing Sels’ 54 qubits, and believes this reinforces their claim of quantum supremacy, even though those findings are pending publication.

Conversely, Sels described D-Wave’s approach as overly defensive, stating, “I don’t see the point of running larger problems just to satisfy them.” He emphasized that the efficiency of their tensor network approach already scales with problem size, making it unnecessary to explore larger calculations.

In addition, researchers Linda Mauron and Giuseppe Carleo from EPFL in Lausanne, Switzerland, also declared that they succeeded in solving the same Ising model problems using classical machines. Their argument aligns with Sels’s, indicating that D-Wave’s 200-year claim for supercomputers to solve certain problems can actually be accomplished in just three days using comparatively modest computational resources.

“The lesson to be learned is that if you say, ‘this is beyond classical simulation,’ then there will be a classical simulation that will do it,” Carleo remarked. He suggested that D-Wave should avoid overstating their capabilities in future publications.

In a rebuttal, D-Wave’s spokesperson dismissed the competing calculations, underscoring that their findings demonstrated simulations being too simplistic to alter D-Wave’s claims. “While this paper does appear to be an advance, it does not challenge our claims whatsoever of beyond-classical quantum simulation,” the spokesperson stated.

Should these skepticism effects prove true, D-Wave wouldn’t be the first quantum computing firm to face criticism over its supremacy assertions. Google previously claimed its Sycamore quantum computer could perform calculations that would traditionally take a supercomputer 10,000 years, only for researchers to later demonstrate the feasibility of those computations using 512 GPUs. Such instances highlight the ongoing debates surrounding quantum computing capabilities.

D-Wave, founded in 1999, has notably evolved from its early days marked by skepticism regarding its unique quantum annealing systems. The company launched its 5,000 qubit Advantage system in 2020, which it touts as the world’s most powerful quantum computer. While there is growing respect for D-Wave, the ongoing dispute about its computational height continues within the wider scientific community.