Banks are not falling behind because of poor execution, but because payments modernisation is still treated as a one-off capital project rather than an ongoing operating discipline. While some institutions are making progress by moving away from episodic transformation towards continuous, expertise-led modernisation, many remain stuck in stop-start programmes that cannot keep pace with regulatory change, operational demands and growing market complexity. David Patrick, Head of Payments Strategy at RedCompass Labs, writes that resilience is built through steady evolution, not periodic overhaul. Applied correctly, AI has a practical role to play, strengthening operational insight, reducing failure risk, and helping payments platforms adapt without introducing fragility.

Payment systems are changing faster than banks can realistically respond.

From ISO 20022 migration to instant payments, cross-border reforms and open banking, the pace of regulatory and technical change has become relentless.

Many banks are now locked in a cycle of reacting, patching, and catching up, and the cracks are beginning to show.

In the US alone, more than two-thirds of banks say they are struggling to keep pace with payments change, and nearly half admit these delays have already triggered outages. In 2025, the European Central Bank went offline for seven hours due to a hardware failure. Citibank suffered a nationwide outage affecting 200 million customers. Nine major UK banks logged more than 800 hours of unplanned downtime in just two years. Global outages now drain an estimated $400 billion a year from the economy, with UK businesses losing £73 million per minute when payments systems fail.

If banks feel like they’re constantly running uphill, it’s because payments modernisation is still approached as a traditional, one-off transformation programme. This is a model that no longer matches the continuous pace of regulatory, operational, and technological change.

The structural challenges slowing modernisation

Banks are not struggling for lack of effort, but because modernisation is still delivered in stop-start programmes, while the payments environment now changes continuously.

ISO 20022 deadlines continue to shift, schemes update rulebooks annually, and cross-border regulators rarely align. Almost half of global banks now cite regulation as their single biggest barrier to progress. Many institutions find themselves rushing to meet minimum compliance requirements rather than delivering value or building long-term capability.

Vendor complexity compounds the problem. Years of tactical purchases (the “buy it when we need it” approach) have created sprawling ecosystems of customised components and incremental fixes. The result is a fragile patchwork that is costly to maintain and difficult to evolve. Sixty-two per cent of banks report spending more than a fifth of their time simply managing customisations, and 85% are considering switching their primary payments platform vendor.

Then there’s the skills gap. More than half of banks have delayed or scaled back programmes, not because of resourcing levels, but because critical payments expertise and platform-specific skills are in short supply. Smaller banks feel this pain most acutely as they struggle to attract specialist payments talent and cannot absorb the financial risk of multi-year transformation programmes. This widening divide means the institutions that most need modernisation are often the least able to deliver it.

Together, these pressures explain why 83% of payments modernisation projects run late, and a similar proportion fail to meet their original objectives.

Why the old approach no longer works

Traditional transformation programmes assume stability. There’s a clear set of requirements, a predictable regulatory roadmap, and a long runway for delivery. Payments no longer operate in that world. Change is continuous.

Legacy systems cannot flex quickly enough. Big-bang replacements are too costly and too risky. Smaller institutions simply cannot afford to experiment with new infrastructure while also keeping day-to-day operations stable.

The result is a modernisation model that is fundamentally misaligned with reality. Payments systems need to evolve constantly, but most transformation programmes are designed to deliver change in large, infrequent chunks. That gap grows every year.

The institutions making real progress are taking a different path, keeping the payments core lean and stable, and surrounding it with modular, interoperable capabilities. They enable continuous, incremental development, strengthen resilience, reduce outages, and improve customer experience without the disruption of major overhauls. But building and running adaptive systems requires two things many banks lack: the right expertise and the capacity to apply it consistently.

Expertise and applied AI: a smarter model for resilience

Banks are investing heavily in AI, particularly in operations, automation, and transaction monitoring. AI is already helping close staffing gaps and speed up decision-making.

But AI alone is not the answer. A late-2025 MIT study found that 95% of enterprise AI projects failed. This isn’t because the technology was flawed, but because organisations tried to use it in isolation. The most successful institutions paired AI with strategic advisors who understood the domain deeply enough to apply the technology effectively.

In payments modernisation, that partnership is essential. AI can accelerate the work by analysing rulebook changes, predicting operational impacts, streamlining testing, and improving auditability, but it must operate within a human-in-the-loop model, where specialist payment expertise provides oversight, judgement, and direction. When combined, they allow banks to modernise continuously, reduce outages and stay ahead of regulatory change.

This is not AI transformation. It’s expertise-led modernisation supported by applied AI, delivered in manageable steps that build resilience rather than destabilise it.

Standing still is now the biggest risk

The pace of payments innovation is only increasing. Modernisation can no longer be treated as a one-off programme. It must become an ongoing discipline.

Banks that keep the payments core lean and stable, and surround it with modular, interoperable capabilities, guided by specialist expertise and strengthened by applied AI, will reduce outages, control costs, and unlock genuine innovation. Those who continue to rely on periodic, disruptive, large-scale programmes will fall further behind.

The institutions that win the next decade of payments will be the ones that keep moving. Standing still is the greatest risk of all.

As payments infrastructure strains under continuous regulatory and operational pressure, these tensions sit squarely along the fault lines reshaping global finance explore more in the latest Trade Treasury Payments magazine issue, Fault Lines, here.