Microsoft's quantum chip Majarona 1 is a few qubits short

Microsoft’s quantum computing scientists announced they have finally realized a long-held goal of building a “topological qubit”, the equivalent of a transistor for ordinary chips, that may help advance quantum computing. 

The qubit is the functional element of a quantum chip, called Majorana 1, based on an exotic particle, a hybrid of matter and anti-matter discovered in the past 20 years.

Also: Google’s quantum breakthrough is ‘truly remarkable’ – but there’s more to do

“Majorana 1 marks a transformative leap toward practical quantum computing,” states Microsoft in the blog post by Chetan Nayak, head of quantum hardware.

However, the caveats suggest you should take Microsoft’s breakthrough with a big grain of salt. 

Also: OpenAI’s o3 isn’t AGI yet but it just did something no other AI has done

For one thing, there’s skepticism about the veracity of Microsoft’s claims, even with a highly technical report in this week’s Nature Magazine by Microsoft’s quantum researchers. 

In addition, the Majorana 1 is not even a beta product as far as computing goes, maybe not even alpha. It cannot do any useful work until it somehow is extended with many more qubits, known in technical terms as “scaling” the device.

The topological qubit is unlike other quantum devices because of the way it leverages a particular breakthrough in materials science.

The “Majorana fermion” is a particle with the property of being both matter and anti-matter. The particle existed as a hypothesis for decades until researchers confirmed its existence in 2012. The name of the Microsoft chip, Majorana 1, is a nod to the importance of the particle.

Microsoft is unique among quantum computing researchers by chasing the fermions as the root of their qubit endeavor. Microsoft claims the topological qubit could be far more useful in any ultimate quantum device because it is, in a sense, more “digital” than other quantum machines. 

Also: If you’re not working on quantum-safe encryption now, it’s already too late

All quantum approaches have to measure the output of a qubit as if measuring a continuous variable like a flow of water. However, the topological qubit essentially alternates between two electrical states, so it’s simpler to measure, as described in a separate technical paper posted by the Microsoft team on the arXiv pre-print server. 

Those elements of Microsoft’s approach are mostly theoretical until proven. And there are grounds for caution because the very existence of a topological qubit is something that has to be inferred, rather than directly observed. 

In the Nature technical paper, Microsoft scientists claim to have indirectly measured the topological qubit inside Majorana 1, though some scientists briefed on the device have their doubts.

Lead author Morteza Aghaee and dozens of colleagues describe the Majorana 1 as a computer chip made of indium arsenide and aluminum. The chip is fashioned as a collection of nanowires that cause a reaction in quantum dots, leading to the topological qubit. 

The presence of the topological qubit is measured by an interferometer, a device that uses the intersection of light beams to measure the “parity” of the quantum dot.

Also: Google’s AI Co-scientist is ‘test-time scaling’ on steroids. What that means for research

However, Nature reporter Davide Castelvecchi notes in an overview article that Aghaee and team are careful to point out that their interferometer’s tests “do not, by themselves” prove there are the necessary Majorana particles in the device. 

And Castelvecchi quotes one physicist, Steven Simon of the University of Oxford, UK, who says, “There’s no slam dunk to know immediately from the experiment” that the qubits are made of topological states. 

Castelvecchi also notes: “Some researchers are critical of the company’s [Microsoft’s] choice to publicly announce the creation of a qubit without releasing detailed evidence.” 

Castelvecchi, who has covered the quantum race for many years, notes that another Microsoft team made a similar claim in 2021 about topological qubits and had to retract it.

Also: Google’s Gemini 2.0 AI promises to be faster and smarter via agentic advances

In fact, this week’s announcement comes many years after Microsoft originally expected to demonstrate a topological qubit. In 2018, in Barron‘s magazine, Microsoft’s quantum team told the paper that the company would have a topological qubit ready by the end of that year. In a sense, the Majorana 1 is seven years late.

That’s not a criticism so much as an example of how quantum in general, and Microsoft’s quest, specifically, takes a lot more effort than people expect based on what’s in a press release.

Which brings us to the more important caveat: Majorana 1 is not a computer chip. It is a prototype of a computer chip. 

Also: What is sparsity? DeepSeek AI’s secret, revealed by Apple researchers

Another thing Microsoft told Barron’s in 2018 is that a functional quantum computer was only five years away. That prediction turned out to be too optimistic.

The Microsoft press piece about the recent achievement states, “Today, the company has placed eight topological qubits on a chip designed to scale to one million.” 

It’s important to note you need the full million to do any real computation. Microsoft insists the million target will be reached “within years, not decades,” but it’s unclear how. 

The arXiv paper has lots of interesting information about expanding the Majorana chip from one qubit to multiple qubits, but it does not make any predictions about time frames.

Who’s to say Microsoft won’t produce a scaled-up Majorana in the coming years? If Microsoft’s claims, so far, can be rapidly extended and validated, then there’s reason to hope progress will be swift in scaling. 

However, without an explicit device roadmap or time frame, Microsoft has the same challenge Google has with the “Willow” quantum chip it unveiled in December: to prove it can be taken from qubit science into a working computer.