The Great Re-Encryption: Why the Government Just Set a Clock on Every Secret It Owns
Adversaries are recording encrypted traffic now to read it when quantum computers arrive. The US just ordered the largest re-encryption in history — and the first test isn't the math, it's a question no agency can answer: where is all our cryptography?
By MyAudioBooks.ai ·
Listen free: The Great Re-Encryption: Why the Government Just Set a Clock on Every Secret It Owns
Somewhere right now, in a data center you will never see, someone is recording encrypted traffic they cannot read. Diplomatic cables, health records, financial transfers, corporate secrets, the classified traffic of governments — vacuumed up in bulk and stored on the shelf, unreadable today, on the bet that a machine will exist someday that can read it. Intelligence agencies have a name for this: harvest now, decrypt later. And it means the quantum threat to encryption is not in the future. Every secret transmitted today with today's encryption is already exposed to the day that machine arrives. The theft is happening now. The reading is just delayed.
That is the threat model the United States government has now formally acted on. On June twenty-second, twenty twenty-six, the White House signed Executive Order fourteen thousand four hundred twelve, and two days later the Office of Management and Budget issued Memorandum M-26-15. Between them, they set the clock on the largest coordinated re-encryption in history: every federal agency must inventory the cryptography inside its most sensitive systems, submit migration plans within a hundred twenty days — plans that come due right about now, in the autumn of twenty twenty-six — and complete the migration to a new generation of quantum-resistant encryption standards for protecting keys by the end of twenty thirty and digital signatures by the end of twenty thirty-one, with full migration targeted by twenty thirty-five. The deadline is not set for the day the quantum computer arrives. It is set for the day before the backlog of stolen traffic becomes readable. And the first test of the entire program is not the mathematics. It is a question almost no agency can currently answer: where is all of our cryptography?
This article is about the strangest infrastructure project the government has ever launched: replacing the locks on every door in the country at once, before the master key exists, because the burglars have already photographed the keyholes. It involves a theorem that breaks half the internet, a decade of intercepted traffic sitting in storage, a brand-new set of mathematical locks that have been standard for barely two years, and an inventory problem so large that the government has had to invent a new kind of parts list just to begin. The story of post-quantum migration is not really about quantum computers. It is about the fact that secrecy has a shelf life, and the milk is already on the shelf.
At My Audio Books dot A I, you can create your own audiobooks from prompts, turn your documents into audio, all with one subscription, and store your items in your own personal library.
Start with the lock that is about to break, because nearly everything runs on it. When you connect to a bank, a hospital, a government portal, almost anything secure, the connection is set up by one of two families of public-key mathematics: RSA, based on the difficulty of factoring enormous numbers, or elliptic-curve cryptography, based on the difficulty of a related problem in curve arithmetic. These problems are chosen because they are easy to compute in one direction and effectively impossible to reverse with ordinary computers — factoring a number with thousands of digits would take a classical machine longer than the age of the universe. In nineteen ninety-four, a mathematician named Peter Shor proved that a sufficiently powerful quantum computer could reverse them quickly — not quickly in theory, quickly in practice. Shor's algorithm is the reason the entire field exists: the day a big enough quantum machine is built, the math underpinning RSA and elliptic-curve encryption stops being a wall and starts being a revolving door. The machine that size does not exist yet — the largest quantum computers today are hundreds to a few thousand error-prone qubits, and breaking production encryption would require something vastly larger and more stable, what specialists call a cryptographically relevant quantum computer. But Shor's theorem does not care about the schedule. It only cares that the wall is temporary.
Now the part that makes the threat present tense. If an adversary records your encrypted traffic today, they cannot read it — the math holds. But the recording does not expire. Health records have privacy horizons of decades; diplomatic and intelligence traffic, longer; a citizen's financial and biometric data, a lifetime. Store the ciphertext, wait for the machine, and every one of those records eventually opens like it was never locked at all. This is why the government's deadline is set by the sensitivity of the data, not the arrival date of the hardware: information that must remain secret past the early twenty thirties cannot be protected with today's locks starting tomorrow — it has to be protected with the new locks starting now, or it is being read in twenty thirty-two the moment it was sent in twenty twenty-six. The harvest is not speculation, either; bulk interception of global traffic is documented practice of multiple state intelligence services, and the storage is the cheap part. The economics are lopsided in the attacker's favor: a warehouse of drives costs less than a single fighter jet, holds a decade of a nation's secrets, and appreciates in value every year the decryption day gets closer. It is the only form of espionage in which the stolen goods improve with age — the rare theft whose value compounds while it sits in the vault. The Executive Order's unusually blunt phrasing frames the migration as a defense against advanced cryptographic attacks — including, implicitly, attacks that have not happened yet but whose raw material is being collected today.
The new locks exist, and they are the other half of the story. After an eight-year global competition that read like the Olympics of mathematics, the National Institute of Standards and Technology finalized its first post-quantum standards in August twenty twenty-four: FIPS two oh three, a key-establishment algorithm called ML-KEM, built from lattices — geometric structures in high-dimensional space where certain shortest-path problems stay hard even for quantum machines; FIPS two oh four, a digital-signature algorithm called ML-DSA from the same family; and FIPS two oh five, a signature scheme called SLH-DSA built on hash functions, kept as a backup in case the lattice family ever cracks. The executive order and the OMB memo lock the federal government onto these standards: ML-KEM for protecting keys by the end of twenty thirty, signatures by the end of twenty thirty-one. Why lattices, of all the mathematics on Earth? Because Shor's algorithm is a specialist, not a generalist. It works by finding hidden periodic structure in the factoring and discrete-logarithm problems — a symmetry that quantum computers can exploit. Lattices hide no such symmetry: the problem of finding the shortest path through a grid of billions of dimensions stays brutally hard for quantum machines too, at least as far as anyone currently knows. That last clause is doing quiet work, and the field is honest about it — lattice problems are hard as far as anyone currently knows, which is precisely why NIST standardized a hash-based fallback in parallel. The architecture of the new standards is itself an admission that certainty is unavailable: one family for daily use, another in the basement in case the first one cracks.
And here is the detail that defines the difficulty of the whole migration: these algorithms are not drop-in replacements. They have bigger keys, bigger signatures, different performance profiles, different failure modes. Swapping them into a system built for RSA is not changing a lock; it is changing the door, the frame, and sometimes the building's wiring.
At My Audio Books dot A I, you can listen to this story and thousands of others that explore the hidden science and mechanics behind the headlines.
Which brings us to the inventory, the part of the story that decides whether any of this happens on time. Before an agency can migrate its cryptography, it has to find it — and that is a much harder problem than it sounds. Cryptography in a modern government is not a switch in a server room; it is embedded in thousands of applications, protocols, libraries, devices, and vendor products, accumulated over decades, documented nowhere. It lives in the VPN concentrator and the SCADA controller, in the diplomatic pouch system and the parking-garage badge reader, in code written by contractors who left in two thousand nine. The OMB memorandum therefore makes the cryptographic inventory the first phase of the whole program: agencies must map their high-value assets and high-impact systems and produce a cryptographic bill of materials — a CBOM, a parts list for encryption, modeled on the software bill of materials that supply-chain security adopted years ago — within two hundred seventy days of guidance from CISA and NIST. The government's own assessments concede the obvious: most agencies do not have this map. The largest re-encryption in history begins with a scavenger hunt for doors.
There is one precedent for an operation like this, and it is worth holding up because people remember it wrong. At the end of the nineteen nineties, the world faced a different cryptographic-adjacent deadline: the year-two-thousand problem, when two-digit years in millions of programs threatened to roll over to zero. The world spent an estimated three hundred billion dollars finding and fixing date logic in every system that mattered, and when the millennium turned, almost nothing failed — which is exactly why everyone now believes Y2K was a hoax. The disaster that does not happen looks, afterward, like a disaster that was never coming. That is the trap the post-quantum migration was born into: if it works, the payoff is an uneventful decade, and the political reward for an uneventful decade is the suspicion that the money was wasted. Every program manager carrying this migration knows the Y2K lesson by heart: success here is invisible, and invisible success is the hardest kind to fund. The difference this time is that the clock is not a date but a threshold — no one can say when it trips, only that everything after it is too late.
One more piece of context belongs before the objections, because it quietly changes the picture: the private internet has already started migrating, and it is ahead of the government. The major browsers have been rolling out hybrid key establishment — classical and post-quantum mathematics run together, so the connection survives as long as either one holds — in ordinary consumer updates over the past two years. The largest encrypted-messaging services shipped post-quantum ratchets years ago, invisibly, to their billions of users. Cloud providers offer post-quantum key exchange at the flip of a configuration flag. The pattern is instructive: the parts of the system that are centrally managed — a browser, an app, a cloud platform — migrated in months, because a small team can flip a switch for billions of endpoints at once. The parts that are distributed, legacy, and owned by nobody — government systems, industrial controllers, hospital devices, the power grid — are the parts that need a presidential order and a decade. Quantum migration is easy where ownership is concentrated and brutal where it is diffuse, and the United States government is the most diffuse owner of cryptography on Earth.
The order's reach also does not stop at the government's own walls, and the ripple is the point. Federal standards have a way of becoming everyone's standards: when NIST finalized the post-quantum suite in twenty twenty-four, it was not writing for agencies alone — it was writing the menu the world's technology industry orders from. The migration mandate multiplies that effect through the supply chain: defense contractors must meet the new baseline to keep selling to the government, and their compliance flows into the same products they sell to hospitals, banks, and utilities. The last time Washington set a cryptographic baseline this way, the algorithm it chose became the default lock for the entire internet within a decade — the Advanced Encryption Standard, selected by the same institute in two thousand one, is now so universal that your processor has a dedicated circuit for it. This is the mechanism by which a memo about federal systems becomes, in practice, a re-keying of much of the private economy as well — not by law, but by procurement, which is stronger than law.
The strongest case against the urgency deserves a full hearing, because serious cryptographers make it and the history of deadlines supports them. Start with the timeline: cryptographically relevant quantum computers do not exist, and the distance from today's thousand noisy qubits to the millions of stable logical qubits that Shor's algorithm needs is not a roadmap but a guess — it could close in five years or in twenty-five, and governments have cried wolf on technology timelines before. Second, the new standards are young: ML-KEM descends from a scheme called Kyber, but the competition that produced it also produced a finalist called Rainbow that was broken on an ordinary laptop before it was ever standardized — a reminder that new mathematics can fail suddenly, and that a government that just bet its infrastructure on a two-year-old standard may yet have to migrate again. Third, the deadline record: the federal government has missed every big cryptographic migration it has ever scheduled — the move to IPv6 is more than two decades late and counting, and the SHA-1 deprecation took years past its deadline. Fourth, the migration itself is the risk: ripping cryptography out of legacy systems breaks things, and the twenty thirties will be full of outages traceable to this memo.
The internet is littered with equipment that hard-coded the old mathematics into silicon a decade ago: middleboxes that inspect traffic, load balancers that terminate connections, embedded controllers in water plants and substations with fifteen-year lifespans and no update mechanism at all. Some of that equipment cannot be migrated; it can only be replaced, and replacement cycles in critical infrastructure run in decades. The memo's deadlines therefore contain a quiet admission embedded in their phasing: key establishment first, signatures second, everything else last — because the parts that can be fixed quickly must be fixed first, and the parts that cannot be fixed at all will have to be isolated behind newer systems rather than upgraded.
The honest version of this critique says the correct priority is not the deadline but crypto-agility — the ability to swap algorithms quickly when one fails, a capability the migration builds either way, quantum or no quantum.
The strongest case for acting now is the milk on the shelf. The traffic being recorded today does not care when the deadline is, and the sensitivity horizon of the most valuable secrets already crosses the twenty-thirty line. Migration takes a decade at federal scale; if you start when the machine exists, you finish a decade after the secrets opened. That is the entire case for the memo, and it fits in one sentence.
Three developments would disprove the skeptics' timeline — or the believers' — and each is observable. First, the physics: if a credible laboratory demonstrates a cryptographically relevant quantum computer at scale — millions of stable logical qubits, or a clear engineering path to them — the skeptics lose the timeline argument overnight, and every deadline in the memo becomes suddenly slack instead of tight. Second, the mathematics: if ML-KEM or its sibling standards take a serious cryptanalytic hit — a structural break like the one that killed Rainbow — the migration's foundation cracks, and the fallback, SLH-DSA and the remaining candidates, becomes the front line overnight. Third, the bureaucracy: if the first reporting cycle shows agencies unable to produce their cryptographic inventories — if the scavenger hunt stalls at phase one — then the entire schedule was fiction from the start, and the real story becomes a government that cannot map its own locks, let alone change them.
It is worth saying what this article has not claimed. It has not claimed a quantum computer can break RSA today; it cannot, and this article has said so precisely. It has not claimed the harvest-now-decrypt-later collection is hypothetical; bulk interception is documented practice, and the storage logic is arithmetic, not conspiracy. It has not claimed the new standards are weak; they are the best-vetted cryptography ever standardized, and youth is not a flaw, it is simply youth. And it has not claimed the federal deadlines will be met; the historical record suggests skepticism, and this article has named it.
Which returns to the data center you will never see, and the shelf of unreadable traffic quietly aging toward the day the locks dissolve. The government's answer is a decade-long project to re-key the country before the shelf empties into the open — a race measured not against a machine's arrival but against the shelf life of secrets already in flight. The strangest part of the post-quantum migration is that its success will look like nothing at all: no ribbon, no launch, no machine to photograph, just a trillion encrypted messages that remain, forever, unread. The deadline is on the calendar now. The harvest is already underway. And the only question the next decade will answer is which shelf life runs out first — the secrets, or the time left to protect them. Somewhere in that same data center, the shelf keeps filling, one quiet packet at a time.
At My Audio Books dot A I, you can create fiction, non-fiction, and turn your documents into audio, all stored in one place with a single subscription — plus get instant access to thousands of audiobooks and deep-dive investigations. Learn more today at My Audio Books dot A I.