The uncomfortable premise of this entire subject is that some encrypted data has already been stolen, and the theft has not mattered yet.
An attacker who intercepts encrypted traffic today cannot read it. But storage is cheap, and patience is free. If a machine capable of breaking today’s encryption arrives in ten or fifteen years, everything captured and archived between now and then becomes readable retroactively.
That strategy has a name — harvest now, decrypt later—and it is the reason a threat that has not arrived yet carries a deadline that already has.
On 22 June 2026, two executive orders were signed setting hard federal timelines for migrating to post-quantum cryptography. The first, Securing the Nation Against Advanced Cryptographic Attacks, reaches beyond federal agencies to contractors and critical infrastructure operators. That reach is what turns a government IT programme into a business story.
Table of contents
- What post-quantum cryptography actually is
- Harvest now, decrypt later, explained properly
- What the executive orders require
- The timeline in one table
- Who is actually in scope
- What a CryptoBOM is and why it comes first
- What not to worry about
- What to do if you run a business
- FAQs
What post-quantum cryptography actually is
Post-quantum cryptography is encryption designed to resist attack by a quantum computer, running on ordinary hardware you already own.
That last clause is the part most coverage buries. PQC is not quantum technology. It does not require a quantum computer, quantum network or any new physics. It is a set of mathematical algorithms — running on the same laptops, phones and servers as today’s cryptography — built on problems that quantum computers are not expected to solve efficiently.
Why replacement is necessary comes down to two families of cryptography and how differently they are threatened.
Public-key cryptography — RSA and elliptic curve, the mechanisms that establish secure connections and produce digital signatures — is the vulnerable part. Its security rests on mathematical problems that a sufficiently large quantum computer could solve dramatically faster than a classical one. This is the category being replaced.
Symmetric cryptography — AES, which does the bulk work of actually encrypting data — is in much better shape. It is weakened rather than broken, and the mitigation is largely a matter of key length.
So “quantum breaks encryption” is too broad. The specific problem is how systems agree on keys and prove identity, and those are the two things the new federal deadlines are organised around.
Harvest now, decrypt later, explained properly
The threat model is what makes this urgent despite the uncertainty, and it is worth being precise because it is routinely overstated in both directions.
An adversary with resources intercepts and stores encrypted traffic now. They cannot read it. They store it anyway. Years later, when capable hardware exists, the archive is decrypted.
Whether that matters to you depends entirely on how long your data stays sensitive.
- A session token that expires in an hour: irrelevant. Nobody will care in 2040.
- A password you will have rotated: mostly irrelevant.
- Medical records, legal files, source code, trade secrets, intelligence material, personal identity data: still sensitive in fifteen years. Highly relevant.
This is the practical filter, and almost no coverage applies it. If your organisation holds data that must remain confidential into the 2040s, harvest-now-decrypt-later is a live concern today. If your most sensitive data has a shelf life measured in months, it is a planning matter rather than an emergency.
TechyKnow’s assessment: the honest position sits between the two loud camps. Nobody can tell you the date a cryptographically relevant quantum computer arrives, and anyone offering one is guessing. But the migration takes years regardless, and the retroactive-decryption risk means the clock started before the machine exists. Both things are true.
What the executive orders require
Two orders were signed on 22 June 2026. The one that matters commercially is Securing the Nation Against Advanced Cryptographic Attacks.
Three provisions carry beyond government:
Hard deadlines rather than guidance. Federal agencies must migrate their most sensitive systems to PQC for key establishment by 31 December 2030 and for digital signatures by 31 December 2031. Previous federal direction on this was strategy; this is a date.
Contractors are in scope. The order directs federal contractors to comply with post-quantum Federal Information Processing Standards by the end of 2030. If you sell to the US federal government, this is your deadline too, and it arrives through your contract rather than through a regulator.
Near-term reporting. Agencies must submit PQC Migration Plans to OMB and the Office of the National Cyber Director by 22 October 2026 — within 120 days of the order. That is the deadline that will generate visible activity this year, and the one likely to reach vendors as questionnaires.
The pattern here should be familiar if you followed the AI governance story: the regulation applies to a defined set of organisations, and then flows downhill through procurement to everyone who sells to them.
The timeline in one table
| Date | What happens |
|---|---|
| 22 June 2026 | Two executive orders signed, including Securing the Nation Against Advanced Cryptographic Attacks |
| 22 October 2026 | Agency PQC Migration Plans due to OMB and ONCD (120 days from the order) |
| End of 2030 | Federal contractors required to comply with post-quantum FIPS |
| 31 December 2030 | Federal agencies migrated to PQC for key establishment on most sensitive systems |
| 31 December 2031 | Federal agencies migrated to PQC for digital signatures |
Verify each against OMB memorandum M-26-15 before relying on them; executive orders and OMB memoranda are frequently amended, and the implementing guidance carries detail the order itself does not.
Who is actually in scope
This is the section vendors do not write, and it will save most readers a great deal of anxiety.
Directly in scope: US federal agencies; federal contractors; critical infrastructure operators reached by the order’s provisions.
Indirectly in scope, and this is the large group: anyone selling software or services to the organisations above. You will meet this as a contract clause or a security questionnaire asking about your PQC roadmap, not as a legal obligation of your own.
Not meaningfully in scope right now: a small business with no government customers, no long-lived confidential data, and infrastructure running on major cloud and browser platforms. Your encryption will be upgraded largely by the vendors whose software you use. Several major platforms have already begun deploying post-quantum key exchange in production traffic, which means some of your connections are already protected without any action on your part.
That last group is most readers, and being told so plainly is more useful than being sold a readiness assessment.
What a CryptoBOM is and why it comes first
Every serious migration guide starts in the same place, and it is not a purchase.
A Cryptographic Bill of Materials — CryptoBOM — is a structured inventory of every encryption dependency across your systems: what algorithms are in use, where, in which products, with which key lengths, and who owns each one.
The reason it comes first is that you cannot replace cryptography you do not know you are running. In most organisations of any age, cryptography is embedded in places nobody remembers: an old VPN appliance, a payment integration, a certificate issued by someone who has left, a library four dependencies deep in an application nobody has rebuilt in three years.
The related concept worth learning is cryptographic agility — designing systems so algorithms can be swapped without re-architecting. The genuine lesson of this transition is not that everyone should adopt one specific set of algorithms. It is that the next transition will happen too, and systems that hard-code cryptographic choices will be expensive again.
What not to worry about
Three things get overstated, and correcting them is part of covering this honestly.
Your bank account is not about to be emptied. No known quantum computer can break production cryptography today. The threat is future capability applied to archived data, not present capability applied to live transactions.
Bitcoin is not about to collapse. Cryptocurrency quantum-doom stories are perennial and typically ignore that protocols can and do change their signature schemes, and that the timelines involved are long. Treat sudden-collapse framing sceptically.
You probably do not need to buy anything yet. The first genuine step is inventory, which is effort rather than expenditure. Vendors offering to sell you a solution before you know what cryptography you are running are selling in the wrong order.
What to do if you run a business
Proportionate to the four groups above.
If you sell to the federal government: treat end-2030 as real and start now. Migrations of this type take years, and the October 2026 agency planning deadline means your customers will begin asking about your roadmap within months.
If you sell to enterprises or regulated industries: expect PQC questions in security questionnaires within the next year or two, following the same downhill path as every other federal security requirement. Having a documented position — even “we track our vendors’ roadmaps and hold no long-lived confidential data” — beats having none.
If you hold data that must stay confidential past 2035: apply the harvest-now filter honestly. This is the one group where the risk is live today rather than scheduled.
If none of the above applies: learn the term, keep your software updated, and let your platform vendors do the work. That is a legitimate answer, and you will not read it on a vendor’s site.
The wider pattern connects to what TechyKnow has covered across agentic AI and infrastructure security: regulation defines a small in-scope group, and procurement quietly extends it to everyone in the supply chain. The organisations caught out are rarely the ones the rule named.
The practical next step: if you are in one of the first three groups, start the inventory. Not the procurement — the inventory. Knowing what cryptography you are running is the prerequisite for every decision that follows, and it is the one task nobody can sell you a shortcut around.
9. FAQs
What is post-quantum cryptography?
Post-quantum cryptography is encryption designed to resist attack by quantum computers while running on ordinary hardware. It is not quantum technology — it is a set of mathematical algorithms based on problems quantum computers are not expected to solve efficiently, replacing public-key systems such as RSA and elliptic curve cryptography.
What does “harvest now, decrypt later” mean?
It describes an attacker intercepting and storing encrypted data today that they cannot yet read, in order to decrypt it once capable quantum hardware exists. It matters because the risk applies retroactively — data stolen now becomes readable later. Whether it affects you depends on how long your data stays sensitive.
When will quantum computers break encryption?
Nobody knows, and specific dates should be treated with suspicion. No known quantum computer can break production cryptography today. The reason deadlines exist despite that uncertainty is that migration takes years and the harvest-now risk applies to data being transmitted now.
What do the June 2026 executive orders require?
Two orders signed 22 June 2026 set federal deadlines: migration to post-quantum cryptography for key establishment by 31 December 2030 and for digital signatures by 31 December 2031, with federal contractors required to comply with post-quantum FIPS by end of 2030. Agencies must submit migration plans to OMB and ONCD by 22 October 2026.
Does post-quantum cryptography affect my small business?
Usually only indirectly. If you have no government customers and no data that must remain confidential into the 2040s, your encryption will largely be upgraded by the platform and cloud vendors you already use. The group that should act now is organisations selling to federal agencies or holding very long-lived confidential data.
What is a CryptoBOM?
A Cryptographic Bill of Materials is a structured inventory of every encryption dependency across an organisation — which algorithms are used, where, in which products, and who owns them. It is the first step in any migration, because cryptography that is not inventoried cannot be replaced.
Is my data at risk from quantum computers right now?
Not from decryption today. The realistic present-day risk is that data being transmitted or stored now could be captured and decrypted years later. That only matters for information which will still be sensitive at that point.
Will quantum computers break Bitcoin?
This is regularly overstated. Cryptocurrency protocols can change their signature schemes, the timelines involved are long, and no current hardware poses a live threat. Treat sudden-collapse framing with scepticism, while noting the underlying concern about public-key cryptography is real.
