Overview
- University of St Andrews physicist Henry Legg published a formal critique in Nature on June 24, 2026 that says Microsoft’s February 2025 paper failed to show the claimed topological gap and that broader data look like random noise.
- Legg identifies specific problems in Microsoft’s tuning software, including coding mistakes that he says hid alternative results and led to inconsistent or misreported outcomes in the transport data.
- Microsoft replied in Nature the same day, defended its methods and said DARPA reviewed its public and proprietary results before advancing the company in a federal benchmarking program.
- The dispute revives questions about reproducibility and research integrity because two earlier Microsoft-backed papers were retracted and editors flagged problems in other work connected to the effort.
- Majorana-based topological qubits aim to store quantum information more stably by using a predicted particle called a Majorana, but independent, widely accepted replication of Microsoft’s core claims has not been published and rivals use different, better-established approaches.