UK finance leaders struggle to prove AI value in cash

FIS has published research showing that many UK finance leaders cannot clearly measure the value AI delivers in cash and liquidity management, even though 90% say they have a clear business case for using the technology.

The survey highlights a gap between adoption and proof of return. While 89% of respondents said they understand where AI can deliver the greatest value and 91% said their teams have the skills to use it effectively, 29% said they cannot clearly measure the business value of their AI initiatives.

Opinium conducted the study among 100 Senior Finance Directors at UK companies with revenue of £250 million or more. It suggests finance teams are under pressure to take greater responsibility for cash and liquidity outcomes while working with fragmented systems and incomplete visibility.

That challenge extends beyond AI. Some 65% of respondents said investment decisions are sometimes made without a complete picture of available liquidity, while 97% said the finance function is becoming more accountable for cash and liquidity results even when the relevant data, systems or processes sit outside its direct control.

Manual work remains a significant factor. Respondents said their finance teams spend an average of 16.2 hours a week gathering, reconciling and validating cash-position data before sharing it with senior stakeholders.

Infrastructure strain

The survey points to structural complexity across large organisations. On average, respondents said they manage 14 legal entities, 11 banking relationships and 10 core finance, treasury, ERP or payments systems, making it harder to maintain a consolidated real-time view of cash.

Only 78% said their systems are sufficiently connected to support AI use cases. A similar share, 79%, said they can access the data needed for AI initiatives without significant manual effort.

Concerns about controls also remain. Some 72% said accuracy, explainability or governance issues are limiting adoption, while 76% cited data security risks.

These responses suggest finance leaders see a practical rather than conceptual problem. The barrier is less about whether AI has a place in treasury and liquidity management than whether data quality and system connectivity are strong enough to support reliable outputs and measurable results.

Asked what would make the biggest difference, respondents ranked greater automation of reporting and reconciliation first, at 46%. A single real-time view of cash across entities and accounts and better system integration were each cited by 43%, while 42% pointed to improved interoperability between systems.

"Finance leaders are investing in AI with a clear sense of where it can add value; however, what we are observing is that the ability to measure that value is not keeping pace with deployment. If you cannot clearly demonstrate what value AI is delivering, it becomes very difficult to build the internal confidence needed to support a move beyond pilots and truly scale. That measurement challenge is closely linked to data quality, trust and connectivity across the enterprise. AI is only as reliable as the information it works with, and in many finance functions, that information is still being assembled manually from disconnected systems. Addressing that requires treasury and liquidity management technology that connects data across banking relationships, legal entities and financial systems. For instance, this interoperability can help reduce the manual effort involved in building a trusted cash forecast and provide AI with the reliable data foundations it needs to support measurable value," said James Mitchell, Global Head of Office of the CFO at FIS.

Data foundations

The results add to a broader debate in corporate finance about whether AI roll-outs are outpacing the data architecture needed to support them. In this case, the survey suggests many finance departments believe in the case for AI but still lack the connected infrastructure needed to verify outcomes and support wider use.

Mitchell added: "Finance leaders don't lack conviction in AI. They often lack the evidence to prove it, and you cannot safely scale what you cannot trust and measure. That measurement gap is really a data quality gap. When a cash position is assembled from disconnected banks, entities and systems, with any manual process involved, neither the finance team nor an AI model can fully utilise it until that data has been rationalised.

As a result, decisions can slow down, opportunities can be missed, and confidence in AI models deployed before that work is done can erode. We expect the organisations that pull ahead to be the ones that address their data foundation and interoperability issues first. That is where FIS can help: our treasury and liquidity solutions are designed to give finance teams a consolidated, near real-time view of liquidity by connecting data across their banking relationships, legal entities and financial systems, helping to reduce manual effort, support a more trusted cash position and forecast, and provide AI with a reliable data foundation to help deliver value leaders can measure and demonstrate."