Microsoft has unveiled a new quantum computing chip redesigned with the help of AI, saying it now believes it can build commercially useful quantum machines by 2029.
The new timeline puts Microsoft on a similar path to IBM, which recently said it plans to spend $10 billion on quantum machines and aims to deliver useful quantum computers by the same year. Microsoft had previously avoided giving a specific target date, saying only that practical quantum systems were years away, not decades.
Microsoft is competing with IBM, Google, Amazon, and several Chinese efforts to develop quantum systems that could one day solve problems in medicine, chemistry, and cybersecurity that would take traditional computers thousands of years to complete.
The company’s new chip, called Majorana 2, follows its first Majorana chip introduced last year. The biggest change is the material used inside the chip. While companies such as Google and IBM commonly build quantum chips using superconducting aluminum wires, Microsoft’s new chip uses lead, a larger atom.
Microsoft said it made the shift with the help of AI tools developed for materials science. According to Jason Zander, the Microsoft executive leading the company’s quantum efforts, the change resulted in a 1,000-fold improvement in some areas of Majorana 2’s performance.
Zander said the key breakthrough was figuring out how to use lead on a chip despite the material being water soluble, which creates major manufacturing challenges. He said Microsoft developed a specialized process that allows the material to be used without washing away during chip production.
Microsoft’s quantum strategy is based on quasiparticles known as Majoranas. The company has claimed to have observed them, but that claim has faced criticism from some physicists who say Microsoft has not publicly released enough data to fully verify its results.
The debate around Microsoft’s quantum claims has continued since earlier studies. Science last year alerted readers that it was investigating data used in a previous Microsoft study from 2020, and some critics say the issues with Microsoft’s data and research methods still remain.
Henry Legg, a lecturer in quantum physics at the University of St. Andrews in Scotland, said Microsoft’s results need to be reproducible, regardless of the materials it uses.
Microsoft executives said they cannot release all of their data because of trade secrets, but said the information has been shared in confidential discussions with the U.S. Defense Advanced Research Projects Agency, which is reviewing different types of quantum systems.
Zander defended Microsoft’s approach, saying the company has done enough physics work to support its engineering investment. He said Microsoft would not spend money on the engineering effort if it still believed the underlying physics was uncertain.





