On June 2, 2026, Microsoft announced Majorana 2 at its annual Build developer conference in San Francisco. The headline claim: qubit lifetimes of at least 20 seconds — and in some cases approaching a full minute — compared to the one-to-twelve millisecond range of its predecessor. That is, by Microsoft's own numbers, a thousand-fold improvement in stability.
The technical story behind those numbers is unusually interesting. Majorana 2 swaps out the aluminum superconductors used in the original chip for lead, which provides better electromagnetic shielding and reduces sensitivity to environmental interference. The fundamental approach — topological qubits based on Majorana particles distributed across nanowires — remains the same. The idea is that because information is spread across the nanowire rather than stored in a single particle, the qubit is inherently more resistant to local disturbances that tend to corrupt conventional qubits.
Microsoft also disclosed that the chip's development was partly accelerated by Microsoft Discovery, its agentic AI research platform. Autonomous agents reportedly automated measurement workflows, optimized fabrication parameters, and flagged performance anomalies — an early example of AI meaningfully shortening the hardware development loop. The company now says it believes a commercially useful, scalable quantum computer could arrive as early as 2029, cutting its prior timeline estimate in half.
That is the optimistic version. The skeptical version, voiced by several physicists publicly, is harsher. Henry Legg, a physicist who has followed Microsoft's quantum work closely, told Scientific American that the data underpinning Majorana 2 come from "a handful of purported instances on a single device" — a sample size that would not survive peer review. "You can see something amazing in one device and never see it again because it's just some random artifact," he said.
The credibility concern runs deeper than this single announcement. In 2021, Microsoft retracted a high-profile Nature paper on topological qubits after outside experts showed the results could be explained by material flaws rather than actual Majorana states. Physicist Sergey Frolov was blunter about the cumulative pattern: "When Microsoft is mentioned these days, physicists and quantum computing specialists just chuckle or raise their eyebrows." A preprint backing the Majorana 2 claims has reportedly been circulating since mid-2025 without landing in a top-tier journal, which critics read as another red flag.
None of this means Microsoft is acting in bad faith. The claims might survive scrutiny — 20-second qubit lifetimes are striking enough that independent replication attempts are already underway. But the gap between what Microsoft announces and what the broader physics community accepts has been a pattern for years, and that gap matters when the company is asking investors, governments, and developers to orient roadmaps around a 2029 quantum computer. The lead-superconductor swap is real engineering work. Whether there is a functioning topological qubit underneath it remains genuinely disputed.