Post-quantum: what actually changes for payments

Quantum computing has moved from a research topic to a planning item. In August 2024 NIST published its first post-quantum cryptography standards, including ML-KEM for establishing keys and ML-DSA for digital signatures. NIST's draft transition plan proposes deprecating today's widely used public-key algorithms, RSA and elliptic-curve cryptography, from 2030 and disallowing them from 2035.

What is at risk

A sufficiently capable quantum computer would break the public-key algorithms that set up secure connections and sign transactions. Symmetric encryption such as AES-256, which protects the data itself once keys are agreed, is considered far more resilient.

So the exposure in payments is less "all encryption fails" and more "the way we exchange keys and prove identity has to change". That touches TLS connections, API authentication, card and wallet provisioning, and code signing.

What does not change

• Data that is decrypted in the wrong place is still exposed, quantum or not.
• Weak access control is still weak access control.
• Logs, backups and test copies holding plaintext are still a problem today.

The property that matters most

The lesson of every cryptographic transition is that it takes longer than planned. Systems that hard-wire one algorithm take years to change. Systems designed for crypto-agility can swap algorithms through policy, without redesigning the applications around them.

The goal is not to pick the algorithm that lasts forever. It is to build systems that can change algorithms without drama.

A useful first step: list where your organisation uses public-key cryptography today, and ask how long each would take to change.