Quantum risk for Irish banks and fintech

Short answer: For banks and fintech firms, quantum computing is a research-stage opportunity in optimization and risk modelling, and a longer-term threat to the public-key cryptography that protects financial data. No cryptographically relevant quantum computer exists today. The practical first step is knowing which data and systems depend on RSA and elliptic curve cryptography.

Two different questions

Quantum computing and finance meet in two separate places. The first is whether quantum algorithms could one day solve financial problems faster. The second is whether a large quantum computer could break the cryptography that banks rely on. They have different timelines and different owners inside a firm, and mixing them up leads to confused planning.

Where researchers see possible uses in finance

A widely cited 2018 review by Orus, Mugel and Lizaso surveys how quantum computing could be applied to finance. It covers quantum optimization, including how quantum annealers could be used to optimize portfolios, look for arbitrage opportunities and support credit scoring. It also discusses deep learning in finance and ways quantum machine learning might improve it.

The review also covers quantum amplitude estimation, which can give a speed-up for Monte Carlo sampling. Monte Carlo methods sit behind many current financial calculations, including derivatives pricing and risk analysis, so a speed-up there would be of direct interest. The authors describe this as an overview of approaches and prospects, and they note that only small quantum processors were available when they wrote it. That is still true in the sense that nothing in routine banking production runs on a quantum computer.

For an Irish bank or fintech, the sensible reading is that these are problems to understand and watch. Optimization, simulation and sampling are the problem classes to know. None of this is a reason to change core systems today.

The cryptographic risk

The second question is more concrete. The public-key algorithms used across payments, online banking, certificates and messaging, mainly RSA and elliptic curve cryptography, would be broken by Shor's algorithm on a sufficiently large, error-corrected quantum computer. Such a machine does not exist yet, and nobody can say with certainty when or whether it will.

The reason to plan early is the harvest now, decrypt later pattern. An adversary can record encrypted traffic or stolen encrypted data now and try to decrypt it years later. Financial records often have to stay confidential for a long time, so the data being protected may outlive the cryptography protecting it.

The standards that exist

NIST approved its first three post-quantum standards on 13 August 2024: FIPS 203 (ML-KEM), FIPS 204 (ML-DSA) and FIPS 205 (SLH-DSA). Falcon was selected for standardization as FN-DSA, and FIPS 206 is still in development. These give vendors a stable target for key exchange and digital signatures.

A practical first step

Start with a cryptographic inventory. List where RSA and elliptic curve cryptography are used, which systems hold data with a long confidentiality life, and which vendors and libraries control the algorithms. Ask suppliers when they plan to support the NIST post-quantum standards. Our separate posts on crypto inventories, ML-KEM and harvest now, decrypt later cover these steps in more detail.

This article is an independent explainer. It is not financial, legal or compliance advice.

FAQ

Is quantum computing used in banking today?

Not in routine production. The review literature describes research-stage approaches to optimization, machine learning and Monte Carlo sampling, and only small quantum processors exist.

Which financial problems do researchers think quantum computers could help with?

The 2018 Orus, Mugel and Lizaso review covers portfolio optimization, arbitrage detection, credit scoring, quantum machine learning, and faster Monte Carlo sampling for derivatives pricing and risk analysis.

Why is quantum computing a risk to financial data?

A large quantum computer running Shor's algorithm could break RSA and elliptic curve cryptography. Data recorded today could be decrypted later, which matters for records that must stay confidential for years.

Has a cryptographically relevant quantum computer been built?

No. Today's devices are small and noisy, and the timeline for a machine that could break these algorithms is uncertain.

What should a bank or fintech do first?

Build an inventory of where RSA and elliptic curve cryptography are used, find data with long confidentiality needs, and ask vendors about support for the NIST post-quantum standards.