Microsoft shows off novel quantum chip that can scale to 'a million qubits'. So far: Eight

Not just a matter of time but a matter of Majorana fermions, too

Updated Microsoft says it has developed a quantum-computing chip made with novel materials that is expected to enable the development of quantum computers for meaningful, real-world applications within – you guessed it – years rather than decades.

Quantum computing firms that saw the price of their shares plummet in January, when Nvidia CEO Jensen Huang suggested quantum computing won't be practical for another 20 years, should welcome Microsoft's forecast, at least. The rest of us can maintain healthy skepticism.

Redmond's boffins on Wednesday announced Majorana 1, a quantum computing chip based on a Topological Core architecture that can – eventually, it is claimed – scale to support millions of quantum bits (qubits) on a single processor.

Unlike digital bits, which have two possible states, one and zero, qubits can be zero, one, or a quantum superposition of zero and one, representing a potentially infinite set of states until measured. Theoretically, quantum computing could enable calculations that would be impractical with classical computers, if existing barriers like the need for quantum error correction can be overcome.

This error correction involves having many times more physical qubits – a two-state quantum system tied to hardware – than the number of logical qubits – high-level abstractions used for programming – being checked.

Commercially useful quantum computers are expected to require millions of physical qubits. We're not there yet. Google's Willow chip, introduced last December, boasts 105 physical qubits. While qubit counts across devices that rely on different techniques aren't directly comparable – IBM's Osprey from 2022 featured a 433-qubit processor – they're still orders of magnitude shy of millions.

But that's what Microsoft claims is now plausible, within a few years.

"Whatever you’re doing in the quantum space needs to have a path to a million qubits," said Chetan Nayak, Microsoft technical fellow in a statement. "If it doesn’t, you’re going to hit a wall before you get to the scale at which you can solve the really important problems that motivate us. We have actually worked out a path to a million."

Currently, Microsoft says it has eight – that's how many topological qubits reside on its breakthrough chip. But there's room apparently for scaling to millions.

Microsoft has been on this path, now marked with a quantum roadmap, for years. Back in 2017, the Windows giant talked up topological qubits at its Ignite Conference.

The Majorana 1 chip relies on what Microsoft calls a Topological Core, meaning its built with a material called a topological superconductor, or topoconductor, that creates a novel state of matter, a topological state that's neither solid, liquid, or gas.

Krysta Svore, Microsoft technical fellow, explains that a hundred years ago, mathematician Ettore Majorana predicted this new topological state of matter and since then researchers have been looking for the Majorana particle, aka Majorana fermion. Now the IT titan says it's found it, and can use it.

"Last year, we were able to observe it for the first time," said Svore. "And this year, we're able to control it and use its unique properties to build a topoconductor, a new type of semiconductor that operates also as a superconductor."

Sounds like a PlayStation 1 game but isn't ... The Majorana. Source: Microsoft

Instead of silicon, the chip is made with indium arsenide, which is used for detecting infrared light and is superconductive under extreme cold. Details can be found in a paper published in the science journal Nature. There's also a paper distributed via Arxiv that elaborates on topological quantum computing.

With the Majorana 1 chip, Microsoft said it can make quantum computers that operate on topological qubits, which require less error correction and are thus more physically scalable than alternative approaches. Instead of needing to create a machine the size of a football field to house millions of qubits, Microsoft's Majorana 1 chip is designed to fit into Azure datacenters.

Microsoft is one of two companies, the other being PsiQuantum, invited by the US Defense Advanced Research Projects Agency (DARPA) to design a quantum computer that's actually useful for industrial applications.

"Our path to useful quantum computing is clear," said Chetan Nayak, Microsoft's technical fellow and corporate vice president of quantum hardware, in an announcement. "The foundational technology is proven, and we believe our architecture is scalable. Our new agreement with DARPA shows a commitment to relentless progress toward our goal: building a machine that can drive scientific discovery and solve problems that matter."

And in just a few years, or so we're told. ®

Updated to add on February 24

Since Microsoft’s announcement, there’s been some skepticism about whether the Windows titan has really achieved the alleged results. Part of the issue is that Microsoft’s paper was submitted to Nature prior to the corporation achieving what’s claimed.

That means that Microsoft’s purported creation of eight topological qubits has not yet been validated by peer review – a creation Microsoft insisted had occurred. The Nature paper actually addresses a more modest achievement, “a device architecture that might enable fusion experiments using future Majorana zero modes,” as the Nature reviewers put it [PDF].

In response to a discussion of the issue on the blog of quantum computing researcher Scott Aaronson, Microsoft researcher Chetan Nayak has attempted to address concerns raised about Microsoft’s claims. Nayak wrote:

Readers of our Nature paper may have noticed that the paper was submitted on March 5, 2024 and published on February 19, 2025. We have continued to make progress in the intervening year. I showed you these new results during our call, and I presented them in detail to more than 100 researchers from across industry and academia at the Station Q meeting this week. I’ll discuss them during my talk at the APS March Meeting.

We have fabricated a two-sided tetron (in the terminology of Phys. Rev. B 95, 235305 (2017)). Both nanowires were tuned into the topological phase via the topological gap protocol, as in Phys. Rev. B 107, 245423 (2023). This is the topological qubit configuration: there are 4 Majorana zero modes (MZMs), one at each end of each topological nanowire.

We have performed both Z and X measurements. These are the basic native operations in a measurement-based topological qubit. In the Z measurement (which measures the fermion parity of one of the nanowires) we see errors occurring on a time scale of ~10 milliseconds, and we attribute them to “poisoning” by above-gap quasiparticles. In the X-measurement (which measures the fermion parity between MZMs on two different nanowires), we see errors on a 5 microsecond time scale, and we attribute them to the overlap between the MZMs at the opposite ends of a wire.

As we said right off the bat, it's good to maintain healthy skepticism.