The people who were fine? The ones who rolled 50 real dice to make their own seed. 128 bits, immune to the bug.
Self-custody isn't just holding the key. It's owning the process that made it.
Plain-language writeup ๐
https://t.co/5X4AOA2lSs
Thieves emptied thousands of Bitcoin hardware wallets in July 2026.
They never broke Bitcoin's cryptography.
They just guessed the keys. Here's how that's possible ๐งต
So the first real deliverable of a migration isn't an algorithm. It's an inventory: every place asymmetric crypto lives, ranked by harvest-now-decrypt-later exposure.
You can't migrate what you haven't mapped โ
https://t.co/7gA5cQ8EtO
Organizations start a post-quantum migration by asking: "which algorithm... ML-KEM? ML-DSA?"
Wrong first question.
The real one: where does asymmetric crypto live in your systems at all? Usually the answer is "more places than anyone wrote down." ๐งต
One app can use RSA / elliptic-curve crypto in ~10 distinct places:
โข TLS + mTLS
โข PKI & signing pipelines
โข messaging prekeys
โข backup / sync encryption
โข JWT & code-signing keys
Each with a different threat model. "Going PQC" isn't one switch.
Why a post-quantum team cares: you can't migrate - or defend - crypto you can't see. The PQC transition isn't about picking ML-KEM. It's inventorying every place crypto lives and how it's configured.
Full write-up โ https://t.co/98lVgtC9x4
We built a scanner to find cryptography a quantum computer will break. Then ran it on 43 popular open-source projects. We expected RSA and elliptic-curve keys - and found plenty. We didn't expect to keep tripping over a much more basic problem ๐งต
TLS certificate verification - turned off. In production. Hardcoded. rejectUnauthorized: false, no way to switch it back on - in the code paths carrying database passwords, SMTP creds, and secret-manager API keys. A bank-vault door, propped open with a brick.
"When will a quantum computer break RSA?", turns migration into a bet on a date nobody can call.
Too early โ recoverable overhead.
Too late โ keys extractable, old traffic decrypted. Unrecoverable.
Bet against the asymmetry, not the date โ
https://t.co/axMO1noZYK
Why we built QuantaKrypto.
Quantum computing was easy to file under "someday." That filing is now wrong.
The standards are finished. Most migrations have not started. Harvest now, decrypt later means the clock already runs.
That gap is the work.
https://t.co/wKZ3dl1pGN
The first tools are live and free. Scan your codebase for quantum-vulnerable crypto in one command:
npx @quantakrypto/qscan ./
Audits, tooling, certification, training. Be quantum-ready before Q-day.
Intro: https://t.co/AelWowNICz
We are building QuantaKrypto: post-quantum cryptography, made tractable.
A quantum computer that breaks RSA doesn't exist yet. But "harvest now, decrypt later" means someone can record your encrypted data today and decrypt it once one exists.
The clock already started. ๐งต
The good news: the hard part is no longer the math. NIST finalized the standards in 2024 (ML-KEM, ML-DSA, SLH-DSA).
The hard part is the engineering: find every place classical crypto lives, migrate without breaking prod, and prove you did.