By Nick Merritt, Executive Director at Designit.
If I was asked to sum up what 2026 has in store for finance, it would be ‘accountability. ‘
In 2025, there was understandably a lot of excitement around the potential of new technologies (looking at you, AI) in the financial services industry. We saw numerous trials, pilots and experiments. Some worked. Some didn’t.
Now in 2026, tech leaders will have the tricky task of demonstrating to fellow board members that these technologies actually deliver ROI.
In a financial landscape that will also be defined by inflationary pressure and increased regulatory scrutiny, this is going to matter all the more. In 2026, every penny will count.
Welcome to the year of “Measurable Productivity”
Over the next twelve months, we are going to see financial institutions shift their investment from “experimental” to “explainable” AI.
Retail banks, for example, will move away from flashy generative assistants and focus on automating the unglamorous but high-volume processes, like onboarding, compliance checks, and internal reporting.
It’ll be much the same story in insurance: the focus will shift towards data accuracy and automation of claims handling.
The point is, AI’s value – or any technology for that matter – depends on what’s going on under the bonnet, not on how glossy the dashboard looks.
CFOs will start to see AI the same way they do any other capital investment – and will scrutinise it accordingly.
Inflation and debt reshape the balance sheet and the workforce that runs it
This scrutiny is necessary, primarily because of the high cost of capital – something that will continue to be the case throughout the year.
With pressure mounting on their bottom lines, firms will be forced to reassess their recruitment strategies.
Retention of top talent will be an issue. High earners will be tempted to relocate to Dubai or Singapore, and this will force companies to reassess the mobility they afford their leadership teams and the incentives they provide for a borderless workforce.
When the AI hits the data wall, financial services will rethink what intelligence really means
The data wall – that is, the point where useful, high-quality data becomes harder to source and more expensive to label – will slow down the rate at which AI models improve. That won’t kill progress, but it will reset expectations.
This might actually be a good thing for financial services. Institutions will turn their attention to domain-specific intelligence; models trained on proprietary, high-quality financial and behavioural data, and away from the pursuit of ever-larger, ever-more-generalised models.
Across the board, internal data will no longer be seen as a compliance headache, but as part of the strategy.
Regulation becomes the defining force in digital finance
As AI becomes embedded in lending, risk modelling, and fraud detection, regulators will begin treating algorithms and model providers as part of the financial system’s critical infrastructure.
Institutions will be expected to prove not only that models are effective, but that they are fair, explainable, and traceable. The PRA and FCA will tighten expectations on model risk governance, auditability, and accountability for third-party vendors. This will drive a new wave of investment in what might be called “governance technology”.
It will feel, at times, like the fun has been taken out of innovation. But in truth, this is the moment where discipline separates genuine advantage from hype.
Making the most of what we’ve got
2026 will not reward innovation as much as it will miserly and accountable operators. However shiny and exciting your tech is, the discipline with which it’s deployed is what really counts.
Efficiency will become the clearest signal of progress. Not abstract promises of transformation, but measurable improvements in cost, speed, resilience, and decision quality. AI, data, and digital tools still matter – but only if their impact is readily apparent and is demonstrable at scale.
By the end of 2026, the winners in financial services will be those who replaced experimentation with execution.
