Quantum computers from IBM and Quantinuum achieved major milestones by tackling classically intractable problems and clearing quantum error correction hurdles. Researchers ran complex circuits up to 70 logical qubits and 55 qubits respectively, demonstrating verifiable computational advantages over conventional systems.
Verifying that a quantum machine genuinely outperforms conventional hardware has long presented a fundamental obstacle in computer science. As quantum systems scale up, checking their results on a classical computer can become prohibitively difficult, requiring calculations that defeat conventional machines unless researchers rely on strong assumptions about internal behavior. Two major demonstrations bypassed this verification bottleneck through entirely different structural strategies, offering clear statistical confidence in their outputs.
Quantinuum Trapped-Ion Systems Test a Mathematical Limit
A research team led by computer scientists Marcello Benedetti and Harry Buhrman of Quantinuum in the UK devised a game based on a computational task called complement sampling to test quantum superposition. In this experiment, potential answers are divided equally into two groups, A and B. A classical computer given a single random answer from group A faces an exponentially increasing difficulty when tasked with returning an answer from group B.
By contrast, Quantinuum’s trapped-ion quantum computers receive the entire set A in superposition. Using what researchers call a swapper circuit, the system transforms the superposition into set B before measurement, winning every round in an ideal setup while the classical advantage shrinks exponentially. Running thousands of different circuits up to 55 qubits, the hardware consistently cleared this mathematically proven classical limit despite hardware noise.
IBM and University of Chicago Execute 70 Logical Qubits in 15 Minutes
At the same time, IBM and researchers at the University of Chicago announced a demonstration operating 70 logical qubits to tackle a problem considered classically intractable. The quantum computation finished in approximately 15 minutes, whereas leading classical simulation methods would face prohibitive runtimes for the same workload.
To overcome verification issues associated with standard random circuit sampling, the team designed a structured alternative that preserves computational hardness while allowing error detection during computation. The encoded design executed 2,415 logical two-qubit operations and 468 logical T gates. Effective logical error rates dropped to 10 times lower than the underlying physical error rates, keeping fidelity unusually high.
“Verification remains one of the biggest challenges in firmly establishing experimental quantum advantage,” said Bill Fefferman, Associate Professor at the University of Chicago. “This experiment develops techniques to better characterize the fidelity of hard quantum states under noise, increasing confidence that the quantum computer is solving a computationally hard problem.”
Bill Fefferman, Associate Professor at the University of Chicago
Soumik Ghosh, a PhD student in Fefferman’s group, noted that advances in verification have the potential to unlock practical applications for forthcoming generations of quantum hardware beyond simply validating experimental setups.
Entering a New Stage of Quantum Advantage
Both experiments point to an evolving operational landscape where raw speed must be paired with error suppression and verifiable confidence. While Quantinuum tested its algorithms up to 55 qubits using trapped ions, the IBM demonstration combined 70 logical qubits with a public release of its circuits and results through the Quantum Advantage Tracker.
“We are now firmly in the quantum advantage era,” said Jay Gambetta, Director of IBM Research and IBM Fellow. “We have demonstrated a quantum computation beyond the practical reach of classical computers that establishes, with statistical confidence, a lower bound on how faithfully it was executed.”
Jay Gambetta, Director of IBM Research and IBM Fellow
These complementary approaches provide developers and businesses with a stronger foundation for trusting quantum systems as they scale toward complex real-world problems.
Related reading