The Quantum-Safe Toolkit: What Singapore's Banks Are Actually Building

A former IBM chief scientist recently argued that Q-Day—the point at which a quantum computer could break much of today's public-key cryptography—may arrive within three to four years rather than decades away. Others believe it will take much longer. Fortunately, the exact date is less important than many assume.

Cryptographic migrations take years, sometimes decades. By the time everyone agrees Q-Day has arrived, the preparation window may already have closed.

The more useful question isn't when quantum computers become a practical threat.

It's with what we replace today's cryptography. If today's encryption is heading for retirement, what comes next?

There are three answers, each solving a different part of the problem. None require a physics degree—just unlearning the assumption that "encryption" means one thing.

The problem, in one sentence: much of today's Internet security—including the systems that authenticate websites and establish secure connections—relies on public-key cryptography built on mathematical problems that are extremely difficult for today's computers but would become vulnerable to a sufficiently powerful quantum computer.

Option One: New Math Problems (Post-Quantum Cryptography)

The most direct solution keeps the same idea—protect data using difficult mathematical problems—but replaces today's algorithms with ones believed to remain secure against both classical and quantum computers.

This is Post-Quantum Cryptography (PQC), and it is the primary direction banks around the world are taking because it works across existing networks without requiring entirely new infrastructure.

The clearest example is the secure connection established whenever a customer logs into online banking. Before any account information is exchanged, the customer's browser and the bank's server perform a cryptographic "handshake" to agree on a temporary session key. Today that handshake relies on algorithms such as RSA or Elliptic Curve Cryptography, both of which would eventually become vulnerable to a sufficiently powerful quantum computer. Many banks and browser vendors are already testing or deploying hybrid handshakes that combine today's algorithms with PQC algorithms, providing protection while the industry transitions.

Behind these new handshakes sit several families of mathematical approaches.

The leading family is lattice-based cryptography. NIST's first post-quantum standards, are built on lattice mathematics, making it the foundation of most current migration efforts. One way to picture it is as a maze stretched across hundreds of dimensions. Finding the correct path is straightforward if you already know the secret, but searching every possible route becomes unimaginably difficult—even for a future quantum computer. [Imagine finding the shortest route through an enormous multidimensional maze.]

Another family, isogeny-based cryptography, once attracted significant attention because it produced remarkably compact encryption keys. Imagine a vast network of islands, where each island represents a mathematical object and only a handful of narrow bridges connect it to neighboring islands. The public key reveals only the island where the journey began and the island where it ended. The private key is the exact sequence of bridges taken to get there. Recovering that hidden route means searching through an enormous maze of possible paths, and no efficient algorithm is currently known for doing so. Importantly, this is a fundamentally different mathematical problem from the factorization and discrete logarithm problems that quantum algorithm solves, so the quantum techniques that threaten RSA and today's elliptic-curve cryptography do not directly apply. [Imagine reconstructing the exact sequence of bridges connecting two distant islands]

Option Two: Physics Instead of Math (Quantum Key Distribution)

The second approach abandons mathematical secrecy altogether and relies instead on the laws of quantum physics.

Quantum Key Distribution (QKD) transmits encryption keys using individual quantum particles of light.

The key idea is simple: any attempt to measure the quantum states carrying the key inevitably introduces detectable disturbances. Imagine a bank's primary data centre exchanging encryption keys with its disaster recovery site across dedicated fibre. If an attacker secretly taps the fibre and attempts to intercept the key while it is being transmitted, those disturbances become visible when both ends compare their measurements. If too many discrepancies appear, the key is discarded and a new one is generated.

Unlike traditional encryption, the security comes from physics rather than computational difficulty.

Singapore has already tested this approach in practice.

Between 2024 and 2025, the Monetary Authority of Singapore (MAS) conducted a proof-of-concept with DBS, HSBC, OCBC and UOB using infrastructure provided by SPTel and SpeQtral to secure communications between the participating banks and the regulator. The published findings demonstrated that QKD could successfully generate encryption keys at operational scale while reliably detecting simulated eavesdropping attempts.

The limitation is equally important. QKD requires dedicated fibre infrastructure between two locations, making it impractical for protecting every system within an organisation. Instead, it is best viewed as protection for selected high-value communication links rather than a universal replacement for existing encryption. Think of it as constructing an armoured corridor for transporting crown jewels rather than replacing every lock throughout the building.

Option Three: The Ability to Change Locks (Crypto-Agility)

The third capability is arguably the most important, even though it is not a cryptographic algorithm at all.

Crypto-agility is the ability to replace cryptographic algorithms without redesigning or rebuilding an entire system. It is not about choosing the perfect algorithm today. It is about ensuring you are never trapped with the wrong algorithm tomorrow. History provides a useful lesson. SHA-1, once widely trusted across banking and the Internet, began showing significant cryptographic weaknesses in the mid-2000s before practical collision attacks eventually led to its retirement. Yet replacing it across enterprise environments took many organisations close to a decade because cryptography had been deeply embedded into applications, hardware and third-party products.

A crypto-agile organisation would replace outdated algorithms far more quickly because the underlying architecture was designed for change from the beginning.

Interestingly, MAS's own sandbox findings highlighted that the hardest challenge was not the cryptography itself. It was securing sustained executive commitment and funding for a migration programme addressing a threat whose arrival date remains uncertain. That challenge will sound familiar to every technology risk professional.

The urgency is not driven solely by when quantum computers arrive. It is also driven by how long cryptographic migrations take. Large financial institutions often require five to ten years to identify, test and replace cryptography embedded across thousands of applications, hardware devices and third-party products. Waiting until Q-Day is confirmed would be like waiting for floodwaters to appear before deciding where to build the levee.

Where This Lands Across the Risk Pillars

Although quantum-safe migration is usually discussed as a cybersecurity topic, it quickly becomes a much broader enterprise risk issue. Technology and Cyber Risk teams must manage the migration itself. Enterprise Risk teams must decide when to invest despite uncertainty over the timeline. Third-Party Risk teams need visibility into whether vendors have credible PQC and crypto-agility roadmaps. Operational Risk teams must minimise disruption as new cryptographic technologies are introduced. Business Continuity teams should recognise that rushed, large-scale technology migrations are themselves a source of operational outages. Quantum readiness is therefore less about cryptography than organisational preparedness.

What Singapore Businesses Should Do Now

• Build a cryptographic asset inventory. You cannot replace the right locks if you do not know where they exist.

• Ask technology vendors about their PQC migration plans and crypto-agility capabilities, then incorporate those responses into your next third-party risk assessment.

• Prioritise systems based on how long information needs to remain confidential, not simply how critical the application is today. Data that must remain secret for fifteen years deserves more immediate attention than short-lived transactional information.

• Begin testing PQC support in non-production environments. Early pilots often reveal compatibility issues with legacy systems and third-party products long before production deployments begin.

• Invest in crypto-agility even before every post-quantum standard has fully matured. The ability to change cryptography efficiently will remain valuable regardless of which algorithms become dominant.

• Continue monitoring MAS publications as the ecosystem develops. The regulator's own trials identified interoperability challenges between different QKD implementations—an important consideration for organisations operating multi-vendor environments.

  
  

Quantum-safe migration is often presented as a race against future technology. In reality, it is a test of organisational preparedness. The organisations that succeed will not necessarily be those that predict Q-Day correctly. They will be the ones that are already capable of changing the locks before anyone needs to test them.