Silicon Valley loves a good timeline. Microsoft is the latest to throw down a gauntlet, claiming its new quantum chip, the Majorana 2, is a staggering 1,000 times more reliable than its predecessor. They say this leap puts them on track to deliver a commercially useful quantum computer by 2029.
It sounds great on paper.
But in the subatomic world, talk is cheap. The reality of quantum computing is a messy, fragile business. It relies on qubits—subatomic particles that can process mind-boggling amounts of data but collapse if you so much as sneeze near them. Microsoft claims its new qubits can survive for an average of 20 seconds. Compared to the milliseconds of the previous generation, that is a massive jump. The company compares it to a phone battery that suddenly lasts for years instead of a single day.
Still, don’t pack up your classical PC just yet.
To actually solve real-world problems, a quantum computer needs millions of qubits. Microsoft’s new chip has exactly 12. You read that right. Twelve.
“We will have a quantum machine in 2029 that can solve commercially viable, reasonable problems,” insists Zulfi Alam, corporate vice president of Microsoft Quantum. It is a bold claim, especially since the company keeps its secret sauce under lock and key, citing commercial confidentiality. We are expected to take their word for it.
For two decades, Redmond has chased a specific, highly controversial approach called “topological” quantum computing. It relies on a quasi-particle first predicted by Italian physicist Ettore Majorana in the 1930s. To coax this theoretical ghost into reality, scientists had to manipulate a completely novel state of matter.
It has not been a smooth ride.
In 2018, Microsoft had to retract a high-profile paper in the journal Nature after its claimed evidence of the Majorana particle evaporated under scrutiny. Scepticism ran deep. Some physicists openly mocked the research, suggesting Microsoft had abandoned science for blind faith.
Microsoft, however, doubled down. They swapped out aluminum for lead as a superconductor, a tweak suggested by human scientists rather than the AI tools they use to speed up research. The result is the Majorana 2.
“We stand behind it 100%,” says Jason Zander, executive vice president of Microsoft Quantum and Discovery. Zander welcomes the debate but urges critics to look at the data. Microsoft has shared its workings with DARPA, the US defense research agency, as part of a validation program. Yet, the paper accompanying this latest announcement has not undergone peer review. Independent scientists are still waiting for hard proof.
If the technology actually works, the implications are massive. We are talking about simulating molecular structures to eliminate microplastics, wipe out forever chemicals, or invent revolutionary fertilizers.
But the engineering hurdles remain absurdly high. Qubits are notoriously temperamental. A tiny temperature fluctuation or a microscopic vibration can ruin an entire calculation. Nobody has built a truly scalable quantum computer yet.
And classical computers are not standing still. Even Google DeepMind’s Demis Hassabis has warned against writing off traditional silicon too quickly, noting that we still do not know the true limits of classical machines.
Microsoft has made a leap, sure. But in the race to build the future, they are still staring across a massive chasm.
