Microsoft's Quantum Chip and the Problem With Believing Press Releases

I'm not a physicist, and anyone writing about quantum computing without saying that up front is either an actual physicist or should be read carefully. But I follow deep tech announcements closely, because the pattern of hype versus reality in that space maps almost exactly onto what I see in AI, and Microsoft's Majorana 1 chip is a clean case study.

The pitch, announced with real fanfare, was a new kind of qubit built on something Microsoft calls a topoconductor, engineered to host Majorana zero modes, which in theory give you a topological qubit that's inherently more resistant to the noise and decoherence that plagues most quantum hardware. Eight qubits on the chip at announcement, with a roadmap talking about a million qubits on a single palm-sized chip eventually. If the underlying physics holds up the way Microsoft says it does, this is a genuinely big deal, because error correction is the single largest obstacle standing between quantum computing and doing anything useful.

Here's the part that made a lot of physicists reach for their skepticism reflexively, and I think reasonably. Microsoft has been here before. A 2018 paper claiming detection of Majorana zero modes was retracted in 2021 after other researchers couldn't reproduce the results and found the original data had been presented selectively. That's not a small asterisk. That's a company's flagship quantum claim getting pulled from Nature after peer scrutiny. So when the same company comes back a few years later with a bigger claim, the correct response from the field wasn't celebration, it was "show us the independently verifiable measurement," and that's mostly what happened.

I don't think the skepticism means the claim is false. I think it means the claim hadn't yet earned the benefit of the doubt, and given the history, it shouldn't have. As of when I'm writing this, more than a year after the original announcement, I haven't seen the kind of broad, independent, peer-reproduced confirmation that would settle this the way the original paper's retraction should have taught everyone to wait for.

My opinion, for what it's worth: I want this to be real. Topological qubits, if they work the way the theory says they should, solve a problem that's currently being solved by throwing enormous numbers of physical qubits at error correction. That would matter enormously. But "I want this to be real" is exactly the emotional state that got the 2018 paper past reviewers who should have caught the selective data presentation, and I'd rather wait for boring, repeated, independently verified results than get excited off a press release a second time.

The lesson generalizes past quantum computing. Any time a company announces a hardware or capability breakthrough with a slick keynote and a roadmap slide, the interesting question isn't how big the claim is. It's who outside the company has independently checked it yet. Usually, for a while, the honest answer is nobody.

FAQ

Is Microsoft's Majorana 1 chip claim fake?

I'm not saying that, and I don't think anyone can say that yet. What I'm saying is that "unverified" and "false" aren't the same thing, and given Microsoft's own 2018 paper got retracted after other labs couldn't reproduce it, unverified is exactly the state this claim is in until independent groups confirm it.

Why does the 2018 retraction matter so much for how we should read the 2025 announcement?

Because it's the same company making a similar claim about the same physics, and last time the claim didn't survive outside scrutiny, it survived selective data presentation that got past peer review first. That's not proof the new claim is wrong, it's just the reason the burden of proof sits with Microsoft this time, not with the skeptics.

Has anyone independently verified the Majorana 1 results by now?

Not that I've seen, and I wrote this more than a year after the original announcement. That gap is the actual story here, not the roadmap slide about a million-qubit chip.