IBM, University of Chicago Achieve Verifiable Quantum Advantage with 70 Logical Qubits

IBM and researchers from the University of Chicago have successfully demonstrated verifiable quantum advantage in a logical computation. The team used 70 logical qubits to execute 2,415 logical two-qubit operations and 468 logical T gates. This achievement exceeds the practical reach of leading classical simulators.

The experiment developed an approach to address a persistent problem in demonstrating quantum advantage: verifying results once computations become too difficult for classical computers to reproduce efficiently. The team used structured sampling instead of random circuit sampling, which allows errors to be detected during computation and retained the same computational hardness criteria.

Bill Fefferman, Associate Professor at the University of Chicago, states that 'verification remains one of the biggest challenges in firmly establishing experimental quantum advantage.' He believes this experiment develops techniques to better characterize fidelity under noise, increasing confidence that the quantum computer is solving a computationally hard problem.

The encoded computation achieved effective logical error rates 10 times lower than the system's physical error rates. The team found that multiple leading classical simulation approaches would require prohibitive runtimes, and the experiment centers on a structured sampling computation without a named scientific or commercial application.