The Laptop That Called Quantum Computing's Bluff
In March 2025, a team of quantum computing researchers published a result in Science with a bold claim attached. Their machine was built from 5,000 qubits — short for "quantum bits." The bits inside an ordinary computer are, at heart, tiny switches: each one is always either off (0) or on (1), and everything a computer does is built from billions of those switches flipping. A qubit is the quantum-mechanical version of that switch, made from things like supercooled circuits or individual atoms, and because it follows the counterintuitive rules of quantum physics, it isn't limited to being just 0 or just 1.
With this machine, the team had calculated the behavior of a fiendishly complex quantum system — something they argued no normal computer could ever do.
It was the kind of "beyond-classical" milestone the industry has spent a decade and tens of billions of dollars chasing. In this field, "classical" simply means any ordinary computer — a laptop, a phone, a supercomputer — so a "beyond-classical" result is proof that quantum hardware can go where all of those machines cannot. Then a small group of physicists decided to check.
What they found, reported in Science on May 21 and amplified across science outlets again in mid-July, is the sort of thing that makes a whole field pause. Researchers at the Simons Foundation's Flatiron Institute, working with collaborators at Boston University, pulled off the "impossible" calculation on regular hardware — and in some setups, they did it more accurately than the quantum computer had. The detail that launched a thousand headlines: much of the initial work ran on a personal laptop.
To see why the problem was considered so hard, it helps to know what makes qubits strange. That ability to be a blend of 0 and 1 at the same time is what physicists call superposition. And when hundreds of qubits become entangled — linked so tightly that you can no longer describe any one of them on its own — the amount of information needed to track the whole system explodes.
"When you have lots of particles that interact by quantum physics, you have this wave function that describes the state of the system," explains Joseph Tindall, the Flatiron research scientist who led the work. "It's this huge object that rapidly gets bigger and bigger the more particles there are." Try to store that object directly on a computer, and you run out of memory almost immediately.
The trick was not brute force but compression. Tindall's team used mathematical structures called tensor networks — which he likens to "a zip file for the wave function" — to squeeze that unmanageably large object into a set of small, interconnected tables of numbers a laptop could actually hold.
To keep up as the entanglement grew, they reached for an even more surprising tool: belief propagation, an approximation method from the 1980s originally built for entirely different problems, like helping artificial-intelligence systems reason under uncertainty. The idea is simple to state: instead of solving a big interconnected system all at once, each piece repeatedly passes its best local guess — its "belief" — to its neighbors, and the guesses are updated until the whole picture settles into agreement. "It's a little more approximate than some of the other methods, but it's way cheaper, and we can run it much more directly on lots of harder problems," said co-author Miles Stoudenmire. Older, more exact techniques, he noted, "wouldn't be able to even start" on three-dimensional problems this large.
The motivation was, by the researchers' own account, partly ornery. "Whenever we see these kinds of claims, we're always a bit skeptical," Tindall said. "Like, 'Did you try this? Did you try that?'" The March paper was a claim of "quantum supremacy" — the field's term for a demonstration that a quantum computer has done something no classical machine could feasibly do — and it became a convenient target precisely because of the size of its boast — "Why not pick this one that has a big claim attached to it?" as Stoudenmire put it. Their results matched theoretical predictions, agreed with the quantum computer wherever it was right, and beat it on accuracy in two of the tested geometries.
It's worth being clear about what this does and doesn't mean, because "laptop beats quantum computer" invites overreach. The Flatiron team did not prove quantum computers are useless. They showed that one specific, heavily publicized "impossible" problem had a clever classical shortcut its original authors hadn't found — and that's how the field is supposed to work. Every supremacy claim is really a dare to classical programmers, and the goalposts move each time someone clears them.
It has happened before. After Google's famous 2019 quantum supremacy announcement, clever classical algorithms later chipped away at chunks of that claim too. What makes the Flatiron result sting is the hardware: earlier rebuttals often needed supercomputers of their own, while this one started on the kind of machine you'd take to a coffee shop.
Tindall and Stoudenmire want no part of a cage match, though. Both stress that the two camps feed each other — classical simulations are exactly what quantum engineers use to check whether their hardware is behaving. "The barrier for entry for us to simulate certain things is a lot easier than for them, because we don't have to build a quantum computer," Tindall said. "I can just write some code and press 'run' on my personal computer." The team is already aiming at harder targets, like simulating moving electrons — the physics behind real materials such as superconductors — where the classical shortcuts may finally run out.
For the quantum computing industry, the timing is awkward. Billions in public and private money ride on the promise that these machines will soon do commercially valuable things no ordinary computer can touch, and every headline about a laptop matching a 5,000-qubit processor chips at that story. The honest reading sits between the hype and the schadenfreude: "quantum advantage" — the milder successor to "supremacy," meaning a quantum computer beats classical ones at a genuinely useful task, not just a contrived benchmark — is probably real and probably coming, but the benchmarks used to sell it deserve exactly this kind of scrutiny. A claim of impossibility is only as good as the classical algorithms nobody has invented yet.
Which is the quietly radical takeaway. The most important instrument in this story wasn't a cryogenically cooled quantum processor humming in a shielded lab — it was a skeptical mindset, a decades-old algorithm, and the willingness to ask whether "impossible" actually meant it. Sometimes the machine that redraws the frontier is the one already sitting on your desk.