A newly published scientific critique is putting Microsoft’s quantum computing research back under the microscope, casting doubt on findings the tech giant has used to support its bold claim that it will have a fully operational quantum computer ready by 2029.
The critique, appearing in the peer-reviewed journal Nature, was authored by Henry Legg, a quantum physics lecturer at the University of St. Andrews in Scotland. His analysis targets a paper Microsoft published in Nature in February 2025 — a study that has become the cornerstone of the company’s entire quantum computing program going forward.
Quantum computers represent a potential leap beyond today’s conventional machines, capable of tackling complex scientific and cybersecurity challenges that current technology simply cannot handle. The field has attracted significant attention from the federal government, with U.S. President Donald Trump’s administration committing $2 billion to quantum research and this week announcing a goal of achieving a scientific quantum system by 2028.
Microsoft is competing in this space alongside tech industry heavyweights including IBM and Alphabet’s Google. However, while those rivals are building machines using more established quantum technologies, Microsoft has spent close to two decades pursuing a different, more experimental scientific path — one it believes could allow it to leap ahead of the competition.
The February 2025 Nature paper at the center of the controversy made a specific claim: that Microsoft had developed software capable of detecting a tiny gap in an otherwise highly conductive wire. That gap is significant because qubits — the basic building blocks of quantum computers — are extremely sensitive and tend to lose their operational state within fractions of a second. Microsoft argues that identifying a stable gap in a conductive wire is a key step toward creating qubits that last longer and perform more reliably.
Legg’s critique, however, challenges that conclusion. He found that Microsoft’s software “yielded inconsistent and misreported outcomes.” He also pointed to a larger dataset that Microsoft made publicly available but did not include in the published paper, saying it showed only random noise with no convincing evidence of the gap the company claimed to have found.
In an interview, Legg used a colorful analogy to describe his concern — comparing the effort to finding an image of Jesus on a piece of toast by searching through an entire bakery’s supply of bread.
“If you’re looking into something which is essentially just random physics, eventually you will find the Jesus in your toast,” Legg said.
Microsoft pushed back in a formal reply also published in Nature, describing the software as a “practical tuning tool” used to identify optimal locations on its chips for placing qubits. Chetan Nayak, who leads Microsoft’s quantum hardware division, told Reuters the software works well enough that the company uses it routinely to configure chips that are now actively performing quantum computing operations.
“It’s almost like arguing, is flight possible or not? And then you’re standing next to an airplane,” Nayak said. “Well, why don’t you hop in and take a ride?”
This is not the first time Microsoft’s quantum research has faced serious scrutiny. Two earlier Microsoft-backed papers were retracted from Nature, while editors issued alerts flagging potential research problems in two additional papers — one in Nature and one in Science. Microsoft has said those previously retracted papers were produced outside its own laboratories and that it did not review the underlying data before they were published.
Microsoft also announced last year that it had identified the Majorana, a long-theorized subatomic particle that is central to its quantum approach. That discovery, however, has not been published in a peer-reviewed journal.
Sergey Frolov, a physicist at the University of Pittsburgh who has previously criticized Microsoft’s quantum research, said the company lacks the track record of reliable experimental evidence that supports the approaches taken by competitors like IBM and Quantinuum — approaches that do not depend on the existence of the Majorana particle.
“Neither Microsoft nor anyone else has laid a foundation where it is clear that these advances are plausible, through a series of reliable experiments,” Frolov said. “On the contrary, we have a series of papers that keep being challenged at the very basic level, by different people.”
