IBM scientists say they've reached "quantum advantage" in three separate experiments, demonstrating that quantum computers can solve problems faster than the world's most powerful classical supercomputers. But here's the catch: the team is actively encouraging other researchers to try disproving their results. This unusual approach reflects a deeper challenge in quantum computing: proving that quantum machines actually work better, not just differently.

On July 28, representatives from IBM, Algorithmiq, Qedma, and the University of Chicago announced the findings at a news conference. Each experiment used IBM's Quantum Heron R3 superconducting quantum computer system paired with novel error mitigation techniques, which help quantum machines produce reliable results despite their notorious sensitivity to environmental interference.

What Makes These Quantum Advantage Claims Different?

The real innovation here isn't just that quantum computers outperformed classical ones. It's that the researchers built in verification mechanisms from the start. "You want to perform computations that outperform classical, right? But you've relied on classical results for the longest time," explained Abhinav Kandala, IBM principal research scientist. "So when you now begin to outperform, or you go beyond classical, how do you know you had the right result?"

"You want to perform computations that outperform classical, right? But you've relied on classical results for the longest time. So when you now begin to outperform, or you go beyond classical, how do you know you had the right result?" said Abhinav Kandala.

Abhinav Kandala, Principal Research Scientist at IBM

This trust problem is fundamental. Quantum computers are extremely prone to errors because they're sensitive to any form of noise, including interference from Earth's magnetic field. When a classical supercomputer like Japan's Fugaku produces a result, physicists can be reasonably confident it's correct. With quantum machines, that confidence has historically been much harder to establish.

How Did IBM Verify the Results?

The team took an unusual approach to build confidence in their quantum results. Rather than simply comparing quantum and classical outputs once, they ran the same experiments across multiple quantum computers and deliberately introduced different levels of artificial noise into each system. This allowed them to test whether their error mitigation strategy actually worked consistently.

• Multi-System Testing: Researchers measured the same circuit on five different quantum computers, including superconducting systems from IBM Boston and IBM Pittsburgh, plus two quantum computers from Quantinuum.
• Controlled Noise Injection: Each quantum computer received different levels of artificial noise and corruption to ensure the error mitigation techniques could handle real-world interference.
• Consistent Output Verification: Despite the different noise levels, all quantum systems produced consistent computational results, demonstrating that the error mitigation approach was reliable.

The three experiments tackled different computational challenges. The first investigated the Floquet transverse-field Ising model, a system physicists use to study how a material's magnetic properties change when driven by external pulses. This problem becomes exponentially harder for classical computers as it scales up, but quantum machines can handle deeper computations because quantum bits (qubits) can exist in a superposition of states, allowing calculations to run in parallel.

The second experiment, conducted by IBM and Algorithmiq, applied the same model to different problems using a different error mitigation method. As researchers scaled up the problem, classical systems began producing inconsistent results, while quantum systems maintained consistency at measured intervals.

The third study took yet another approach. Researchers designed circuits using "Clifford gates," which are intentionally easy for classical computers to simulate. Then they progressively made the circuits harder by injecting more difficult "T gates." This design guaranteed that error mitigation would work at complexities beyond what classical supercomputers could handle.

Why This Matters: The Verification Challenge in Quantum Computing

Quantum advantage claims have become somewhat routine over the past few years. Google, IBM, and other labs have all announced breakthroughs in "quantum supremacy," "quantum utility," or "quantum advantage." But most of those achievements were eventually surpassed by improved classical algorithms or more powerful supercomputers.

The problem is that physicists cannot imagine every possible mathematical method for conducting classical computations when they test quantum computers. A classical computer scientist might discover a clever new algorithm that solves the same problem faster than previously thought possible, effectively erasing the quantum advantage.

IBM's approach of inviting scrutiny is designed to address this uncertainty. "The classical back-and-forth will keep going on, I think," Kandala noted. "And that should; that's how science progresses. And that's precisely why we have the Quantum Advantage Tracker, a benchmark for measuring quantum advantage. A lot of these problems have been on the tracker for a while now, and I'm sure getting the papers out will get more eyes on it".

"The classical back-and-forth will keep going on, I think. And that should; that's how science progresses." said Abhinav Kandala.

Abhinav Kandala, Principal Research Scientist at IBM

The research papers describing all three experiments were uploaded to arXiv, a preprint server where scientists share findings before peer review. This public release means the scientific community can immediately begin examining the claims and attempting to find classical computing methods that might replicate the quantum results.

How to Understand Quantum Advantage Claims

• Verification Matters More Than Speed: The key innovation in IBM's approach is building verification mechanisms into the experiment design, not just demonstrating that quantum computers are faster.
• Error Mitigation Is Critical: Quantum computers are inherently noisy; the real breakthrough is developing techniques that produce reliable results despite environmental interference.
• Reproducibility Across Systems: Running the same experiment on multiple different quantum computers and getting consistent results provides stronger evidence than a single measurement.
• Expect Ongoing Challenges: Classical computer scientists will continue developing new algorithms to solve these problems, so quantum advantage is not a permanent achievement but a moving target.

The implications extend beyond academic bragging rights. If quantum computers can reliably solve problems like chemical reaction simulations in minutes instead of years, that could accelerate drug discovery, materials science, and optimization problems across industries. But first, the field needs to establish that these advantages are real and reproducible, which is exactly what IBM and its partners are attempting to demonstrate.