Expert again criticizes Microsoft’s advances in topological quantum computing
One of the central challenges in the field of quantum computing, often overestimated, remains the difficulty in objectively evaluating performance and new developments, as much of it depends on indirect measurements. This complexity is particularly evident in topological quantum computing, which employs Majorana fermions. In recent years, Microsoft’s quantum computing division, known as Azure Quantum, has announced significant advances in this area, which have been repeatedly challenged in peer reviews.
The company’s latest attempt to demonstrate progress in this type of quantum computing received a sharp rebuttal from Henry F. Legg, in a paper published in the journal Nature, which brought to light lingering questions about the validity of these claims. This scenario highlights the intrinsic difficulty of validating discoveries in the field, a crucial point for the advancement of science.
Microsoft had already faced scrutiny in early 2025 when it claimed to have detected the fundamental Majorana Zero Mode (MZM), before being the target of severe criticism in reviews from other researchers, including Legg himself. Academic discussions even included terms such as “essentially fraudulent” to describe some of the approaches presented by the company’s team.
This repetition of challenges raises the question whether Microsoft researchers working in quantum computing would be excessively motivated to confirm a discovery or whether there would be a less problematic explanation for the results presented.
The differences between Majorana and Dirac
The unitary operation associated with anion exchange is based exclusively on braid topology, a fundamental concept in quasiparticle physics.
More on this story: IBM’s quantum advance: company machines outperform classical computers in tests
In the traditional approach to quantum computing, Dirac fermions are generally employed as qubits for calculations. However, this methodology has proven to be quite complex and full of obstacles, such as decoherence and noise interference, which make long-term calculations difficult. This requires repeating operations multiple times so that the error correction algorithms have a chance to produce a reliable result.
In this context, topological quantum computing emerges as a promising alternative. Although it presents some limitations in its feature set, it would be considerably more resistant to external interference. It is important to avoid confusion with the reference to Majorana particles, as fermions can manifest themselves as Dirac, Majorana or Weyl. What we are referring to here is, in reality, a Majorana anion, a quasiparticle that shares the property of being its own antiparticle, just like Majorana fermions.
By uniting these anions with braid theory, using the intertwining of their “world lines”, it becomes theoretically viable to perform operations that could be used in the construction of a topological quantum computer.
At its core, this technique would replace trapped quantum particles, which are highly unstable, with braided Majorana anions, which are significantly more robust. If this approach is confirmed, it could represent a revolutionary milestone in the world of quantum computing.
How to prove the existence of Majorana?
Even after creating a device that, in theory, should generate Majorana anions, the next big challenge lies in confirming its effective existence. This is, broadly speaking, the sore point where Microsoft’s attempts have consistently failed in recent years, as evidence is obtained through indirect means rather than through direct and simple measurements or experiments.
Learn more: Helios quantum computer achieves unprecedented precision and accelerates technology evolution
When the first semiconductor transistor was demonstrated at Bell Laboratories in 1947, in the form of the world’s first point-contact transistor, the feat was the result of many years of theoretical study and fruitless attempts, beginning around the 1920s.
At that time, the evidence of a working transistor was unquestionable, as the device clearly operated as a current amplifier. Basic current and voltage measurements at the time were sufficient to prove its effectiveness. The design was later commercialized before being replaced by the bipolar junction transistor and other devices that emerged after the basic principles were demonstrated.
In the current scenario of quantum processors, whether traditional or topological, there is no obvious method to replicate such an elementary demonstration. Even quantum annealing, a simpler and commercially offered approach by D-Wave, is shrouded in controversy over the real existence of a “quantum advantage”. This is a domain where even the mighty IBM has seen its claims of quantum superiority challenged and surpassed by researchers using a humble Commodore 64.
In relation to Majorana anions and the MZM proofs, it is possible, naturally, to seek the construction of a finished device that demonstrates a clear quantum advantage, or to develop a more restricted device, where the existence of these fundamental elements is inferred based on assumptions that remain largely theoretical.
See Also
Singer Robbie Williams reveals autism diagnosis and finds relief in new perspective
Hulk scores with a header and leaves everything the same in the first half of Fluminense x Palmeiras in the Brasileirão
NASA’s New Roman Telescope Sets to Uncover Billions of Galaxies and Dark Energy
In their most recent attempt to prove the creation of these anions and, consequently, topological superconductors, the Microsoft team employed a new method called the Topological Gap Protocol (TGP). This protocol would supposedly allow a parity reading to be performed on its manufactured devices, using it as proof that, this time, the objective had been achieved.
The strong response in peer review
Legg’s latest criticism came as a direct response to the Microsoft Azure Quantum paper, published in Nature, which presented the results of this new approach. The Microsoft paper stated that with the TGP-enhanced test setup, topological qubits had been detected. This conclusion was based, again, on indirect measurements and analysis of recorded data. However, in his criticism, Legg questioned exactly this analysis of measurements, accusing it of having been carried out incorrectly.
The main problem pointed out by Legg is the selective interpretation of measurements, focusing only on data that corroborated the experiment’s hypotheses, configuring, in essence, a confirmation bias. He also argued that Microsoft researchers made several errors in their Python code by using the array’s index instead of its value. After correcting these basic programming mistakes, Legg obtained completely different results from the same measurements.
On the same topic: Artificial intelligence and quantum computing drive advances in the discovery of materials of the future
Legg’s findings indicate that coding artifacts can significantly affect transport-based topological gap detection.
As highlighted by Legg, it is possible to obtain very similar data patterns from other sources, such as quantum dots. Coupled with the issues surrounding data processing, which are considered fundamental, this raises substantial doubts about how close the Microsoft team actually came to creating these topological qubits.
Microsoft defends itself against the accusations
The Microsoft system model displays energy spectrums and the arrangement of ports for the interference circuit, central elements in the company’s proposal.
The Microsoft team promptly responded (albeit behind a paywall) to the scathing criticism. Its main arguments suggest that the Topological Gap Protocol (TGP) plays no role in the interpretation of the radiofrequency results, which are the basis of the initial conclusions, and the company also does not recognize the validity of the problems with the TGP highlighted by Legg.
Another point raised by Microsoft is that Legg did not present an alternative physical model capable of reproducing the capacitance signal or the RTS phenomenology, which, according to the company, sums up Legg’s answer to a simple and dry “no, no”.
Despite recognizing a small misalignment error of a pixel in TGP processing, the Microsoft team insists that this is a minimal error, with no significant impact on the results.
More on this story: New era of quantum computing: Chinese ‘Jiuzhang 4.0’ breaks optical speed record with 3,050 photons
In practice, the criticism is rejected by Microsoft, which maintains the validity of its original 2025 paper. This would imply that topological qubits have, in fact, been detected and that, with this knowledge, a functional topological quantum processor could be developed and integrated into larger systems.
Science continues to search for answers
Although the academic environment and science, in general, can sometimes seem like a stage for clashes where the parties involved exchange barbs, the scientific method must always prevail. This implies the dissemination of results, experimental configurations and methodologies in sufficient detail so that other researchers can reproduce the findings based on their fundamentals.
If Microsoft researchers are correct, this moment could represent a historic milestone for the “point contact transistor” in the world of quantum computing. Such a feat would be quickly confirmed by other teams, who would build their own devices and carry out their own tests, solidifying the discovery as a historical fact.
However, in recent years we have witnessed the tragic end of the Korean superconductor LK-99, which operated at room temperature, and the controversial EmDrive, both of which were repelled by peer review. Cold fusion, meanwhile, remains in ongoing limbo, despite now being referred to as “low-energy nuclear reactions,” exemplifying the scientific community’s wariness in the face of major announcements.
Perhaps the greatest reward of science is that, even if a line of research does not produce the expected results, it still offers a fascinating opportunity to deepen knowledge in physics, mathematics and many other areas. Ultimately, this makes even a topic as controversial as topological quantum computing an enriching subject to explore periodically.
