Table of contents
- An eroding assumption
- More than a research timeline
- AI is not just optimisation
- AI has changed the clock
- PQC will not finish in time
- Start at the pipes
If your quantum-security plan is built around 2035, it is already obsolete. That date may still matter for compliance. It should no longer govern risk.
Until recently, the working assumption was that breaking today’s public-key cryptography would require quantum computers so large that the threat still felt comfortably distant. Depending on the model, estimates were often in the many-millions-of-qubits range.
That assumption is now eroding.
Google Quantum AI, Atom Computing, IBM, Microsoft and neutral-atom researchers announced significant news recently: resource estimates for using Shor’s algorithm against 256-bit elliptic-curve cryptography (Elliptic Curve Cryptography, ECC) are going down, while hardware roadmaps are increasing the number of qubits on quantum computers.
As a reaction, Google and Cloudflare already moved their post-quantum cryptography migration timeline to 2029 – the Ethereum Foundation followed with the same target year.
This Is No Longer Just a Research Timeline.
This is the part boards need to understand: Q-Day is not moving closer because of one announcement. It is moving closer because several independent trends are reinforcing each other:
- The size of the quantum computer thought necessary to break ECC is shrinking.
- The hardware roadmaps are accelerating.
- Neutral-atom architectures are changing assumptions about scale.
Major players are publicly moving their own migration targets for post-quantum cryptography (Post-Quantum Cryptography, PQC) to 2029.
And the algorithmic optimisation process is becoming social, open, competitive, and AI-assisted.
But AI is not just about optimisation.
Recently, Anthropic published research demonstrating how its Claude Mythos Preview model found a materially improved attack against a candidate in the US National Institute of Standards and Technology’s (NIST) process for standardising additional post-quantum digital-signature schemes. Mythos found the weakness after approximately 60 hours of work.
Now it is important to be precise about what it does – and does not – mean.
This does not mean that PQC has been broken.
It was a candidate algorithm, not a deployed NIST standard.
However, the real significance of the Anthropic research is not that one candidate PQC algorithm was weakened. Cryptographic candidates have been weakened or eliminated before. That is precisely why serious algorithms undergo years of adversarial review.
The significance is the speed and scale at which this review can now occur.
AI is becoming a powerful tool for defenders. It will help researchers test algorithms more extensively, examine obscure cryptographic implementations and identify weaknesses before systems are deployed.
But the same capability will be available to attackers.
Attackers need hours. Infrastructure owners need years.
AI has changed the clock.
We just watched this movie with the evolution of AI. For years, progress looked linear. Then models, infrastructure, tooling, data, capital, open-source communities, and automation loops all started reinforcing each other.
Timelines collapsed.
Quantum cryptanalysis is beginning to show the same pattern.
The prudent conclusion is not “Q-Day is definitely in 2029.”
The prudent conclusion is: 2029 must now be treated as the outer planning horizon for protecting your most critical quantum-vulnerable links.
That is roughly 3.5 years.
PQC Migration Will Not Finish in Time.
For a large enterprise, 3.5 years is not enough time for full PQC migration.
PQC migration is not a patch. It is not a certificate rotation. It is not a library upgrade.
It requires cryptographic inventory, Public Key Infrastructure (PKI) modernisation, Hardware Security Module (HSM) upgrades, certificate-chain redesign, protocol testing, embedded-device replacement, vendor coordination, application refactoring, regression testing, auditability, and phased deployment across business-critical systems.
For a large organisation, realistic PQC migration is a 10–15 year programme.
That creates a gap.
A compliance gap.
And, most importantly, a protection gap.
For critical data-in-motion, the answer cannot be: “Wait until every application, service, endpoint, vendor product, embedded system, and certificate chain has been migrated to PQC.”
That will not happen fast enough.
Start at the Pipes.
The practical answer is to start at the pipes.
Protect the critical communication links first:
- Data-centre to data-centre.
- Data-centre to cloud.
- Cloud interconnects.
- Backbone routes.
- Financial trading links.
- Defence and critical infrastructure links.
- Sovereign data flows.
- High-value R&D networks.
- Board and executive communications.
This is where Quantum Key Distribution (QKD) becomes strategically important.
The question is not “QKD or PQC?”
The answer is PQC plus QKD.
PQC is essential for broad cryptographic modernisation and regulatory alignment.
But QKD gives you something PQC does not: an out-of-band keying layer for the most critical links.
QKD can provide umbrella coverage for all traffic crossing a critical link.
QKD is available today. In field-proven, stable, telco-grade, European solutions.
One QKD-secured interconnect can protect many applications, including legacy systems, bespoke internal services, cloud workloads, and vendor platforms that will take years to migrate individually.
For critical links, this can be easier, faster, lower-risk, and more cost-effective than trying to force immediate PQC coverage across the entire enterprise stack.
Begin PQC migration. But protect your critical links now.
If your organisation has communication paths where compromise would be existential, those links should already be on a QKD implementation roadmap. Q-SAFE Solutions can help you identify them, prioritise them, and protect them.
