Last month, Microsoft announced at Fanfare that it had created a new kind of problem and used it to create a quantum computer architecture that could lead to a machine. It can solve industrial-scale problems that have meaning over many years, not decades“.

But since then, the tech giant has been increasingly burning from researchers who say it’s not doing something of a kind. “My impression is that the response of the expert physics community is overwhelmingly negative. Personally, people are just furious.” Sergei Frolov at the University of Pittsburgh, Pennsylvania.

Microsoft’s claim is based on an elusive, exotic quasiparticle called Majorana Zero Modes (MZMS). These can theoretically be used to create topological kibits, new types of qubits, i.e. components of information processing within quantum computers. Due to their unique properties, such qubits can be excellent at reducing errors and can address the major drawbacks of all quantum computers used today.

MZM is theorized to emerge from the collective behavior of electrons at the edges of thin superconducting wires. Microsoft’s new Majorana 1 chip contains some such wires, and according to the company it contains enough MZM to create eight topological maize. A Microsoft spokesperson said New Scientist Chip was a “big breakthrough for us and the industry.”

However, researchers say Microsoft does not provide sufficient evidence to support these claims. In addition to the press release, the company published its paper in the journal Nature He said the results confirmed the results. ” Nature The papermark shows a peer-reviewed confirmation that not only did Microsoft have been able to create majorana particles, but it also helps protect quantum information from random interference, but also allows for reliable measurement of information from that information. A Microsoft press release said.

But the editor Nature It explicitly made it clear that this statement was incorrect. A published report on the Peer-Review process states, “The editorial team wants to point out that the results of this manuscript do not represent evidence of the existence of Majorana Zero Mode in the device on which it was reported.”

In other words, Microsoft and Nature They are directly contradictory to each other. “The press release says something completely different [than the Nature paper]” I say Henry Legg At St Andrews University, UK.

This is not just an unorthodox aspect of Microsoft’s papers. Legg points out that two of the four peer reviewers initially gave rather critical and negative feedback. The peer review report shows that by the final round of editing, one reviewer still opposed the publication of the paper, and three others registered with it. spokesman for Nature I said New Scientist The ultimate decision to publish it came down to the possibilities we saw for future experiments with MZM on Microsoft devices.

Also, one of the reviewers is rare. Hao Chang Legg says that at China’s University of Tsingea, previously collaborated on MICSOFT and MZM research. The work published in Nature In 2018, it was later withdrawn, and the team apologized, “.” Scientific rigor is insufficient” After other researchers have identified inconsistencies in the results. “That’s very shocking Nature You can choose the judge who retracted the paper just a few years ago,” says Legg.

Chang says there was no conflict of interest. “I wasn’t an employee at Microsoft either. [the firm]. Of the more than 100 authors of Microsoft Paper recently, I have worked with three before,” he says. “It was seven years ago, but back then they were Tu Delft students. [in the Netherlands]not an employee of Microsoft. “

Microsoft says the team wasn’t involved in the selection of reviewers and was not aware of Zhang’s participation until the review process was completed. Nature The decision was based on a spokesman who said, “The quality of the advice received can be seen from the reviewer’s comments.”

Looking at the issue, both Leg and Frolov are making more fundamental challenges to Microsoft’s methodology. Experiments using MZM have proven extremely difficult to perform over the past decades. This is because imperfections and obstacles within the device can produce false signals that mimic quasiparticles even if they are not present. This was a challenge for researchers related to Microsoft, including the withdrawn 2018 paper. The withdrawal notice explicitly refers to new insights into the impact of the failure. To address this, Microsoft has been working on 2023. The procedure has been published in the journal Physical Review b It was called the “Topology Gap Protocol” and claimed to tease these differences.

“The whole idea of this protocol was that it was a binary test of whether Mallorna is there,” says Legg. His Unique analysis of code and data However, Microsoft implemented the protocol in 2023, which showed that it was less reliable than expected and changing the format of the data is sufficient to turn the failure into a path. Legg says he raised these issues with Microsoft before its publication. Nature Paper, yet the company was using protocols in new research.

NatureA spokesman for the journal’s editorial team “are aware that some people are questioning the effectiveness of the topology gap protocol used.” Nature Paper and other publications. This was an issue that we were also aware of during the peer review process. “Through the process, the reviewer determined that this was not an important issue at the end of the day, the spokesman said.

Microsoft says it will respond to leg analysis of the 2023 paper. Physics Review B. “Criticism can be summarised as a leg that will build a false strooger for our paper and attack it,” said Microsoft’s Chetan Nayak. He challenged some points to Legg’s work, saying that the 2023 paper “showed that we can confidently create topology phases and Mayorana Zero modes,” and the new paper only strengthens those claims.

A Microsoft spokesperson said: Nature The paper was submitted for review and the company built on its confidence and not only created multi-kut chips, but also tested how to operate these kitz as needed for a working topological quantum computer. The company will release more details at the American Physics Society’s Global Physics Summit in March, the spokesman said. “We look forward to sharing our results and transforming our 20+ year vision of quantum computing into a concrete reality, along with the additional data behind science.”

But for Frolov, the assertion that incomplete results from the past can be ignored as the company is trying to build a more sophisticated device lies in false logic. Legg shares this view. “The fundamental issues of obstacles and materials science don’t go away just because we start manufacturing more fancy devices,” he says.