St Andrews physicist says Python bugs sink Microsoft's Majorana quantum claims

Dr. Henry Legg, a lecturer at the University of St Andrews, published a peer-reviewed critique in Nature on 24 June 2026 arguing that Microsoft’s headline topological-quantum results don’t hold up. The dispute centers on Microsoft’s 2025 Majorana work and its Topological Gap Protocol (TGP), the method the company used to claim it had observed and controlled Majorana particles—the basis for its prediction of a ‘meaningful quantum computer’ within years, reinforced this month by a ‘Majorana 2’ device touted as 1,000 times more reliable.

Legg’s case rests on two coding flaws in Microsoft’s analysis software. The plotting code hardcoded a filter (zbp_cluster_numbers=[1]) that surfaced only one region of data; widening it to [1,2] exposed additional results the original chart never showed. A second routine reversed arrays by index (x[::-1]) rather than by the actual physical bias-voltage values, distorting how the data was read. Together, Legg argues, the code cherry-picked evidence supporting Microsoft’s thesis while hiding contradictory signals, and he frames the company’s reasoning as circular—capacitance measurements alone don’t prove the underlying physics, which he suspects reflects disorder rather than genuine Majorana behavior.

Microsoft is standing firm. Technical fellow Chetan Nayak defended the results and roadmap, dismissing the issue as a ‘minor off-by-one-pixel bug’ with no real impact and saying alternative readings of the transport data fail under more rigorous analysis. The stakes are high: Legg puts functional topological quantum computing ‘centuries, not decades’ away, and notes that Majorana 2—like its predecessor—has yet to publish the X-measurement validation tests that proved troublesome the first time around.