TL;DR. The encryption protecting most internet traffic today, including your website, your email, and your VPN, relies on maths that a powerful enough quantum computer could break. That computer does not exist yet, but the threat is real and already active: adversaries are collecting your encrypted data now to decrypt it later. The solution exists. NIST finalised three post-quantum cryptography standards in August 2024.[1] Google, Apple, Signal, and Cloudflare have already deployed them. In the UK, the NCSC has set a clear deadline: full migration by 2035, with high-priority systems by 2031.[13] This article explains what is happening, what it means for your business, and what to do next.
What Is Quantum Computing
A normal computer stores information as bits, either a 0 or a 1. A quantum computer uses qubits, which can be 0 and 1 at the same time (called superposition) and can be linked together in ways that let them explore many possibilities in parallel. This does not make quantum computers universally faster. But it makes them exceptionally fast at a specific class of mathematical problems, including the ones that underpin most of today’s encryption.
IBM has already built processors with over 1,000 qubits.[2] That sounds like a lot, but breaking real-world encryption would require millions of stable, error-corrected logical qubits. Nobody knows exactly when that becomes possible. Most credible estimates say somewhere between 10 and 20 years. Some put it sooner. Nobody says it is impossible.
What Breaks and What Does Not
In 1994, mathematician Peter Shor showed that a quantum computer could efficiently solve the mathematical problems that RSA and elliptic-curve cryptography (ECC) rely on.[3] RSA and ECC are the foundations of HTTPS, TLS certificates, digital signatures, and most key exchange protocols in use today. If a large enough quantum computer arrives, all of those break.
Symmetric encryption, like AES, is in a different position. In 1996, Lov Grover published an algorithm that would speed up brute-force attacks on symmetric keys.[4] But the effect is roughly halving the effective key strength. AES-128 becomes equivalent to a 64-bit key, which is a problem. AES-256 becomes equivalent to 128 bits, which is still considered safe. So if you are already using AES-256 for stored data, you are in reasonable shape. If you are using AES-128, you should upgrade.
The Risk That Already Exists Today
You do not need a quantum computer to have a problem right now. Adversaries, particularly nation-states, are harvesting encrypted traffic today with the intention of decrypting it once the technology matures. This is called the “harvest now, decrypt later” attack, and the NSA has explicitly documented it in its Post-Quantum Cybersecurity Resources.[5]
Any data that will still be sensitive in 10 to 15 years is already at risk. That includes legal contracts, intellectual property, client records, employee data, and anything covered by long-retention compliance requirements. If you send or store information that would hurt your business if it became public a decade from now, the window for protecting it is now.
The Solution: Post-Quantum Cryptography
Post-quantum cryptography (PQC) means cryptographic algorithms that are safe against both classical and quantum computers. They do not require a quantum computer to run. They work on normal hardware today, just using different mathematics.
In August 2024, after an eight-year evaluation process, NIST published three finalised standards.[1] The actual standard documents are published at the NIST Computer Security Resource Center.[6]
ML-KEM (FIPS 203): Module Lattice Key Encapsulation Mechanism. This replaces the key exchange step in TLS (today done with ECDH). “ML” stands for Module Lattice, a class of mathematical structures that quantum computers are not expected to solve efficiently. ML-KEM-768 is the recommended general-purpose level. Its public key is about 1.2 KB, compared to around 256 bytes for RSA-2048. Slightly larger, but already deployed at scale by Google and Cloudflare.
ML-DSA (FIPS 204): Module Lattice Digital Signature Algorithm. This replaces ECDSA for digital signatures, covering TLS certificates, code signing, and document signing. Signatures are around 3.3 KB. RSA-2048 signatures are 256 bytes. The trade-off is size for quantum resistance.
SLH-DSA (FIPS 205): Stateless Hash-Based Digital Signature Algorithm. An alternative signature scheme based purely on hash functions rather than lattice maths. More conservative design. Slower and produces larger signatures, but the underlying maths is simpler to verify and trust.
For encryption at rest, AES-256 is already considered quantum-resistant. The main changes are in key exchange and digital signatures.
Practical Examples
- Your HTTPS connection today uses TLS with an ECDH key exchange. After a PQC migration, it uses ML-KEM-768 instead. The user experience is identical.
- A software vendor signs their releases today with RSA-4096. After migration, they sign with ML-DSA-65. The installer verification process looks the same.
- Your cloud file storage encrypts files with AES-256. This already meets the post-quantum bar for symmetric encryption. No change needed there.
- Your VPN client and server negotiate a session key using ECDH today. After migration, both ends need updated software that supports ML-KEM.
Where Encryption Lives in Your Business
Most SMEs use encryption in more places than they realise.
Transport encryption (TLS/HTTPS) covers your company website, your SaaS tools (CRM, accounting software, project management), API connections between your services, email in transit, and remote access. Every connection that starts with https:// or uses a VPN is relying on the key exchange and certificate infrastructure that needs to migrate.
Digital signatures cover software you install (code signing), PDF documents with legal signatures, email signing (S/MIME or PGP), and any document verification your business relies on.
Encryption at rest covers your cloud storage (Google Drive, S3, Dropbox), database encryption, device encryption on laptops and phones, and backups. Here, the question is just whether you are using AES-256 or something older.
Authentication infrastructure is also in scope: SSH keys for server access, API tokens, and certificate-based login. RSA and ECC SSH keys will need replacing.
The Vendors Are Already Moving
This is the genuinely encouraging part. The largest platforms have not waited.
Google shipped hybrid PQC key exchange (X25519Kyber768) in Chrome in August 2023, protecting TLS connections for billions of users by default.[7]
Apple deployed PQ3 for iMessage in February 2024, replacing the previous encryption with a protocol designed to resist quantum attacks.[8]
Signal deployed PQXDH in 2023, upgrading its end-to-end encryption to resist harvest-now-decrypt-later attacks.[9]
Cloudflare enabled post-quantum key agreement by default across their network, meaning any website behind Cloudflare now benefits from PQC-protected transport automatically.[10]
Bruce Schneier, one of the most respected voices in cryptography, has written extensively about the importance of migrating now rather than waiting for quantum computers to arrive.[11]
For most small and medium businesses, the practical message is this: if your website runs behind a major CDN, if you use mainstream SaaS tools, and if your email goes through a major provider, a lot of the heavy lifting is being done for you. But you still have gaps, particularly in your VPN, certificate renewal planning, and any custom software or APIs your business relies on.
What AI Has to Do With This
AI affects post-quantum readiness in both directions.
On the positive side, AI tools can scan your codebase and flag every place where RSA, ECDH, AES-128, or MD5 appears. This kind of automated crypto inventory would take weeks manually and minutes with the right tool. AI can also help generate migration plans, suggest compliant replacements, and check new code for outdated algorithm choices.
On the negative side, AI coding tools are a real problem here. They are trained on years of existing code that uses RSA-2048, ECDH, and AES-128 because those were the standards when most training data was written. Left unconfigured, AI code generators will produce code with pre-quantum cryptography defaults. If your team is using any vibe-coding workflow today, there is a reasonable chance your new code is being written with cryptographic choices that will need replacing in a few years. The fix is the same as covered in our previous post: add explicit cryptographic requirements to your AI tool configuration so the model defaults to AES-256, ML-KEM, and ML-DSA from the start.
Why Waiting Is a Problem
The UK position. The NCSC has published a clear three-phase migration timeline for UK organisations.[13] By 2028, organisations should have identified which of their cryptographic services need upgrading and built a migration plan. From 2028 to 2031, they should be executing high-priority upgrades. From 2031 to 2035, the goal is completing migration across all systems and products. The target date for completing the transition is 2035. The NCSC Annual Review 2025 describes PQC migration as “an international, multi-year technology change programme” and dedicated a full chapter of its annual review to the topic.[14]
In January 2026, HM Treasury published the G7 Cyber Expert Group’s roadmap on PQC transition for the financial sector. It aligns with the 2035 overall target and specifically identifies 2030 to 2032 as the window for migrating critical systems. The roadmap explicitly flags the challenge for smaller institutions with higher levels of vendor dependency, which applies directly to most SMEs.[15]
US and international requirements. In December 2022, the US Office of Management and Budget issued Memo M-23-02, requiring federal agencies to begin inventorying their cryptographic dependencies and plan migration to PQC.[12] The NSA’s Post-Quantum Cybersecurity Resources set out migration timelines for national security systems, with some categories required to be PQC-ready by 2030.[5] If your business has US customers or operates in regulated sectors that follow US government procurement requirements, these timelines affect you too.
GDPR. If a breach occurs and it emerges that your encryption was known to be vulnerable and you took no steps to address it, your ability to demonstrate “appropriate technical measures” under Article 32 weakens considerably. The harvest-now-decrypt-later threat makes this relevant today, not just in 2035.
Operational reality. Cryptography is embedded deep in software stacks, certificates, and infrastructure. You cannot migrate it over a weekend. The organisations that start their inventory now, and know what they use, where, and on what timeline it expires, will be in a manageable position. Those that ignore it until the threat becomes front-page news will be migrating under pressure, at higher cost, and with higher risk of mistakes. The NCSC guidance is explicit that the 2035 deadline assumes organisations start now, not in 2030.
Conclusion
Quantum computing will not break your encryption tomorrow. But the threat is real, the timeline is uncertain, and one version of the attack (harvest now, decrypt later) is already active. The standards are finalised, the big vendors are deploying, and the tools to migrate exist.
For most small and medium businesses, the practical steps are these. Audit where encryption is used in your business: websites, VPNs, APIs, storage, and signing. Check whether your stored data uses AES-256 (if not, upgrade). Ask your CDN, email provider, and SaaS vendors what their PQC migration plans are. Flag cryptographic dependencies in any custom software you run. If your team uses AI coding tools, configure them explicitly to use post-quantum defaults.
The NCSC expects UK organisations to have a migration plan in place by 2028. That is less than two years away. The first step, a cryptographic inventory, is something you can start this month.
You do not need to understand the maths. You need to know that the migration is coming, that UK regulators have set a timeline, and that the businesses who plan ahead will have a much easier time than those who do not.
We help SMEs identify where post-quantum migration is needed and build a practical plan to get there.
Get in touch →References
[1] NIST, "NIST Releases First 3 Finalized Post-Quantum Encryption Standards," August 13, 2024. nist.gov
[2] IBM, "IBM Quantum," IBM announced the 1,121-qubit Condor processor in December 2023 as part of its quantum development roadmap. ibm.com/quantum
[3] Shor, P.W., "Algorithms for Quantum Computation: Discrete Logarithms and Factoring," Proceedings of the 35th Annual Symposium on Foundations of Computer Science, IEEE, 1994. PDF
[4] Grover, L.K., "A fast quantum mechanical algorithm for database search," Proceedings of the 28th Annual ACM Symposium on Theory of Computing (STOC), 1996.
[5] NSA Cybersecurity Directorate, "Post-Quantum Cybersecurity Resources," National Security Agency. Includes CNSA 2.0 migration timelines and guidance on harvest-now-decrypt-later risks. nsa.gov
[6] NIST Computer Security Resource Center, "FIPS 203: Module-Lattice-Based Key-Encapsulation Mechanism Standard," August 2024. FIPS 204 and FIPS 205 are linked from the same page. csrc.nist.gov
[7] Chromium Blog, "Protecting Chrome Traffic with Hybrid Kyber KEM," August 2023. Describes deployment of X25519Kyber768 for TLS key exchange in Chrome. blog.chromium.org
[8] Apple Security Research, "iMessage with PQ3: The new state of the art in quantum-secure messaging," February 2024. security.apple.com
[9] Signal, "The PQXDH Key Agreement Protocol," 2023. signal.org
[10] Cloudflare Blog, "Cloudflare's post-quantum cryptography goes GA," 2023. Documents the rollout of default PQC key exchange across Cloudflare's network. blog.cloudflare.com
[11] Schneier, B., Schneier on Security. Schneier has published extensively on post-quantum migration urgency and the NIST standardisation process. schneier.com
[12] US Office of Management and Budget, "Memorandum M-23-02: Migrating to Post-Quantum Cryptography," December 2022. Requires federal agencies to inventory cryptographic systems and plan PQC migration. whitehouse.gov
[13] National Cyber Security Centre (NCSC), "Timelines for migration to post-quantum cryptography," ncsc.gov.uk. Sets out the UK three-phase migration timeline: inventory by 2028, high-priority upgrades 2028-2031, full migration by 2035. ncsc.gov.uk
[14] National Cyber Security Centre (NCSC), "Migrating to post-quantum cryptography," NCSC Annual Review 2025, Chapter 3. Describes PQC migration as "an international, multi-year technology change programme." ncsc.gov.uk
[15] HM Treasury / G7 Cyber Expert Group, "Advancing a Coordinated Roadmap for the Transition to Post-Quantum Cryptography in the Financial Sector," January 2026. Sets 2035 as overall migration target and 2030-2032 for critical systems. Notes specific challenges for smaller institutions with high vendor dependency. gov.uk