Microsoft's Claimed Breakthrough: Majorana 1

February brought significant news from Microsoft, who has long pursued a novel approach to scalable quantum computing based upon a theorized topological qubit. This article assesses what Microsoft has actually achieved, and what is likely press-release hype.

Summary

Microsoft has likely advanced the state of the art in nano-device fabrication and measurement of Majorana Zero Mode evidence, but there remains scientific community contention around these results.

They are maintaining relationships with non-topological qubit platform partners for reasons claimed to be advancing Quantum Error Correction (QEC) and algorithm development, but more likely as a continuing hedge-bet in case they are unable to deliver a qubit.

Microsoft has almost certainly not developed a topological qubit at this time, and has not materially demonstrated a "direct path" to 1,000,000 or more qubits.

Important Reminders

If you are an Upshift Intelligence client, you have heard me insist that achieving quantum computing scale is unlikely to result from a single scientific or engineering breakthrough. A quantum computer is a complex stack of technologies that span from subatomic interactions to new types of circuit devices to many layers of sensing and manipulation equipment to developer tools and cloud services. Advancements must be made at each of these layers; an unlock in one can reveal a bottleneck in another. We must keep this in mind as we evaluate claims of direct paths to millions of qubits.

For a portion of my career, I helped Microsoft establish its quantum computing platform. This included leading teams that developed the Quantum Development Kit (QDK) which enables developers to code algorithms in a quantum-friendly way. Shortly thereafter, we launched Azure Quantum, a cloud service designed to democratize the varying quantum architectures by making them available to developers everywhere through a consistent interface (featuring ion traps, superconducting, neutral atom, and other quantum system from a variety of companies). This gave me valuable insights through collaboration with some of the world's leading quantum labs - both academic and industrial - as we integrated expertise and technology. My perspectives come from this experience. I am not a degreed physicist nor mathematician. For analysis of quantum measurements, I rely on a strong network of thought leaders, many of whom were consulted in the development this brief.

Anyone interested in the advancement of compute capacity for humankind should be ready to celebrate unlocks in these areas. Each press release, and every set of new claims, should be met with excitement and support... and followed by a thorough review of the substance. I endeavor to offer one such review here.

What Was Announced

On February 19, 2025 Microsoft published an update from their quantum program: "Microsoft’s Majorana 1 chip carves new path for quantum computing." This article states several important advancements:

• A new "Topological Core architecture" powering a "quantum chip" called Majorana 1.
• "Topoconductors," a purported materials breakthrough enabling more reliable and scalable qubits. It is described as an entirely new state of matter, not solid/liquid/gas, but topological in nature.
• "A hardware-protected topological qubit" supported by an article published in the journal Nature.
• A "a clear path to fit a million qubits on a single chip." This suggests that only predictable, solvable engineering problems stand between Microsoft and 1,000,000 qubits, not scientific / research unlocks that happen on Mother Nature's timeline. This is described as being on path in, "years, not decades."
• Microsoft's inclusion in the Defense Advanced Research Projects Agency (DARPA) program to evaluate paths to scalable quantum computing through fault-tolerant quantum systems.
• Ongoing partnerships with both Quantimuum and Atom Computing to explore what is possible with "today's qubits."

The announcements were spread across many releases targeting different audiences. These include:

• Microsoft Source - the company blog. A post by Catherine Bolgar shared the news for the broadest possible audience. Note that Catherine Bolgar is not a member of Microsoft's quantum team; she is an author who does "corporate storytelling."
• The Azure Quantum blog - from the quantum team itself. Dr. Chetan Nayak, a credible scientist and leader in Microsoft's quantum program, is named as the author representing his talented team of theorists, device designers, and quantum control and readout experts.
• The journal Nature - a technical publication home to countless announcements of scientific progress over the years. An article titled, "Interferometric single-shot parity measurement in InAs–Al hybrid devices" describes the measurement technique and experimental results observed in achieving the purported goal.
• Pre-Print Research - A device roadmap paper submitted to Cornell University's archive of scholarly articles. The paper has not received significant peer review.
• Satya Nadella's LinkedIn post - sharing the news and congratulating the team. It suggests that a "nearly 20 year pursuit" has concluded with this announcement.

What Is Likely Real?

Microsoft has very likely refined its materials development, fabrication, and quantum sensing (measurement) techniques to become more precise. They have been at this for over a decade, and they have advanced the field in partnership with university and industrial labs historically. Most of that has been brought in-house, and the measurement claims are likely valid. The inferences from those claims get further scrutiny below.

Related to the above, it is likely that Microsoft's measurement techniques have been refined, and that measurement errors reduced. "Less than 1%" was claimed in the technical papers, and merits some celebration for improving the state of the art of this type of measurement.

What's Real but Old News?

Microsoft developed a protocol that they used to validate observation of topological gaps requiring superconducting states and the presence of Majorana Zero Modes (MZM). The measurements that they report are consistent in the materials used in previous publications: indium arsenide (InAs) and aluminum (Al), or InAs-Al. These results were shared back in 2023 in the American Physical Society's Physics Review Journals. This announcement met with some debate, which continues. The claim is similar to the current announcement, and has generated similar debate from the community. Prior to that, an August 2017 paper in the same journal Nature had to be retracted by Microsoft based upon community invalidation of the findings; the track record is far from perfect.

DARPA's Underexplored Systems for Utility-Scale Quantum Computing initiative has indeed selected Microsoft - along with PsiQuantum - as targets to benchmark quantum systems. It is interesting that the architectures powering these two platforms are very different: Microsoft's topological, PsiQuantum's photonic. There is some logic to the Microsoft selection, since they have recently shared positive progress on the quantum error correction front, but that was in partnership with Quantinuum using ion-trap systems, not their own theorized topological system.

The DAPRA agreement and advancements in device fabrications and measurement are worth the respective companies celebrating, though the true metric for excitement would come via published findings from DARPA. I strongly suspect those will never be made public.

What's Not Real?

I will assert three truths that have sufficient supporting evidence:

• Microsoft has not [yet] created a topological qubit.
• Microsoft does not have a clear path to 1M qubits in years.
• Microsoft does not have a chip capable of quantum computation.

Those are some bold assertions, so let's look at the factors that led me to these conclusions:

First, a very close read will uncover very carefully worded depictions of Microsoft's actual state, especially from the more reputable sources who are less likely to play into press hype. For example, Dr. Chetan Nayak was thoughtful to say, "rapid advancements on the path to useful quantum computing," and, "ability to harness a new type of material and engineer a radically different type of qubit." Dr. Nayak is a trained physicist and computer scientist, so his words are likely to be the most accurate (after the headlines, anyway).

If you compare this to the language used by executives who are not trained in these fields, you see much more hyperbole. This tense exchange on LinkedIn is between an accomplished physicist and Zulfi Alam. Zulfi runs Microsoft's quantum program as a Vice President. He is not a physicist; he previously managed the team that built gamepad controllers for Xbox (I know, because I managed the team that built the platform software for Xbox). Zulfi is a well-intentioned and I wish him much success in his role, but I recognize that he is trying to keep a team funded. The thread demonstrates the critical eye that the physics community is casting on the announcement. It is warranted based upon earlier similar claims from Microsoft being discredited and retracted: (above, below)

It is common for there to be healthy debate in the physics and materials communities around announcements of this magnitude. If you paint a large target on your chest with headlines, many arrows will soon arrive. That is what we see happening here. The challenge Microsoft faces is: the marksmen firing the arrows are well-accredited critics with highly informed arguments countering the claims, and very little firm evidence to shield Microsoft from the arrows.

If you paint a large target on your chest with headlines, many arrows will soon arrive.

There are three final pieces of evidence that close my arguments of what's not real:

First comes from Microsoft itself: the acclaimed Majorana 1 chip supposedly powered by a topological qubit is not accessible to anyone (I'd argue not accessible to Microsoft itself). For several years, the Azure Quantum cloud service has played host to a spectrum of qubit platforms under the auspices of giving the public access to them to fully characterize them and begin developing algorithms for them. If Majorana 1 was as conclusive as Satya's posts suggest, it would be resolutely allaying questions about the measurement data via a public appearance on Microsoft's own accessible quantum platform. This reinforces Microsoft's perceived path, rather than an actual topologic qubit device, much less an 8-qubit device as was later inferred.

Second, we see poignant questions from authoritative insiders with both academic and industry credentials. Dr. David Reilly, former lead scientist of Microsoft's quantum control and measurement platform, world-renowned professor of quantum computing at University of Sydney, and leader of a stealth quantum startup, had this to say about Microsoft's claims of a path to "a million qubits fit on a chip," when interviewed by Financial Review:

"A lot of qubits range in size when you're looking at just the nanoscale part, but when you zoom out a little bit they all look very similar in terms of the wiring and the interconnect and the control systems," he said.

"Yes size matters, but it's a complex trade-off between size and functionality and I'm not sure anyone at this point can definitively say how these trades should be optimised."

Third and finally, is Nature's own commentary on the submitted paper:

"The editorial team wishes to point out that the results in this manuscript do not represent evidence for the presence of Majorana zero modes in the reported devices."

The very journal into which this result was published for critique by the scientific community refutes fundamental assumptions inferred from the measurement results, much less the presence of a qubit, much less a highly error-protected qubit, much less a chip full of them, much less a quick path to millions of them on a chip.

In Conclusion

Good work continues to be done at Microsoft. The attention-grabbing headlines are being churned out not because they have a topological qubit, but because one of their chief competitors, Google, announced substantial progress with their superconducting qubits last year.

Microsoft's quantum program has promised a topological qubit for well over a decade. The company's support for the expensive program is stressed without that result. When competitors lead in press coverage, Microsoft throttles up to compete in the headlines where it can't compete with qubit-based systems. That's mostly what we see happening here.

One thing is certain: the deluge of headlines with bold claims has us talking about whether or not Microsoft has achieved a topological qubit. We're completely ignoring the more important question: when will topological protection be demonstrated, and will a topologically-protected qubit be substantially better than any other qubit architecture when finally realized?

If not, how many years - or decades - will it take to improve it to the theorized accuracy? More on this in the next brief.