By Stephen Nellis
SAN FRANCISCO, June 24 (Reuters) – A new critique in the scientific journal Nature is raising fresh questions about Microsoft’s claimed quantum computing breakthrough last year, which underpinned the company’s announcement this month that it will have a working quantum system by 2029.
Quantum computers could solve scientific and cybersecurity problems beyond the reach of conventional machines. They have become a priority for U.S. President Donald Trump’s administration, which invested $2 billion in the field and this week set goals for a scientific quantum system by 2028.
Like Big Tech rivals IBM, Alphabet’s Google and others, Microsoft is developing its own quantum computer. But while rivals are engineering machines based on better-understood quantum technologies, Microsoft has spent nearly two decades trying to break new scientific ground on a technology it says could help it leapfrog competitors.
In a formal reply to the critique and an interview with Reuters, Microsoft said it stands behind its research and that its quantum program is making practical progress despite any concerns.
Microsoft’s scientific effort has drawn skepticism. Two previous Microsoft-backed papers were retracted from Nature, while editors flagged alerts about possible research problems in two others, one in Nature and another in Science.
Microsoft said the previously retracted papers in Nature were done outside its labs and it did not review the data in them before publication.
The peer-reviewed critique published in Nature on Wednesday by Henry Legg, a lecturer in quantum physics at the University of St. Andrews in Scotland, raises concerns about a fifth paper, published in February 2025, and an associated press announcement. The paper, which is not being retracted, is central to all of Microsoft’s subsequent quantum efforts.
Microsoft said publicly last year it had found the Majorana, a long-theorized subatomic particle central to its approach. However, it has not published that discovery in a peer-reviewed journal, such as Nature. The February 2025 Nature paper made a narrower claim: that Microsoft had developed software to identify a minute gap in an otherwise highly conductive wire.
The gap matters because qubits, the basic units of quantum computers, are powerful but fragile, often losing their state within fractions of a second. Microsoft says that finding a stable gap in a conductive wire is part of a process that could create longer-lasting, more useful qubits.
Legg, however, found Microsoft’s software “yielded inconsistent and misreported outcomes.” He also said a broader dataset Microsoft released but did not include in the paper showed random noise, with no clear evidence of the gap Microsoft claimed to find.
In an interview, Legg compared the effort to finding an image of Jesus in toast by looking through an entire bakery’s worth of loaves.
“If you’re looking into something which is essentially just random physics, eventually you will find the Jesus in your toast,” Legg said.
In its reply in Nature, Microsoft defended its claims and said the software was a “practical tuning tool” to find good places on its chips to place qubits.
Chetan Nayak, who oversees Microsoft’s quantum hardware efforts, told Reuters in an interview that the code works well enough that Microsoft regularly uses it to set up chips now carrying out 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?”
Sergey Frolov, a University of Pittsburgh physicist who has also criticized Microsoft’s work, said Microsoft lacks the longstanding evidence supporting the approaches taken by rivals such as IBM and Quantinuum that do not rely on the existence of the Majorana.
“Neither Microsoft nor anyone else has laid a foundation where it is clear that these (Majorana-based) 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.”
(Reporting by Stephen Nellis in San Francisco; Editing by Peter Henderson and Christopher Cushing)
Comments