Finance leaders are being asked to modernise at speed while tightening their grip on cash. Protiviti’s latest Global Finance Trends Survey shows how closely those two demands have become entwined. AI is now used by 77% of finance organisations, while economic, monetary and trade policy volatility has pushed liquidity and cash management towards the front of the workload. 

Cash has forced its way to the front of that agenda. Some 83% of CFOs and finance leaders rank cash management among the three areas requiring the most attention because of market developments. Among CFOs specifically, 47% name cash management as the single area demanding the most attention, ahead of FP&A, cost optimisation and contingency planning. 

Investment in technology therefore faces a practical test. Finance needs better visibility over cash, earlier warning when forecasts move, and enough flexibility to respond before working capital or funding comes under pressure.

Security and privacy of data remains the number one finance priority for the third consecutive year, followed by financial planning and profitability analysis, enhanced analytics, process improvement and strategic planning. AI has leapt from 13th to sixth overall and from 14th to fourth among CFOs and vice-presidents of finance. The technology is being absorbed into an agenda still dominated by familiar finance problems: planning, profitability, data and process efficiency. 

Volatility puts cash back in command

Economic uncertainty is reaching directly into routine finance work. Across financial reporting, forecasting, profitability and sourcing, roughly three-quarters of organisations report at least a moderate impact from changes in economic, monetary and trade policies. 

Reliable reporting and forecasting are particularly exposed. Some 64% say their ability to prepare reports and forecasts within required timelines has been moderately affected, with a further 15% assigning a higher impact score. Among privately held companies, 89% report at least a moderate effect. 

Profitability and the ability to source goods are also under pressure. Three-quarters of respondents report at least a moderate impact in that area, including 15% describing the effect at one of the two highest levels. 

Policy volatility helps explain why cash has moved so decisively up the agenda. Across all respondents, cash management is the most commonly ranked first priority arising from policy volatility, chosen by 37%. FP&A follows at 23%, cost optimisation at 15% and contingency planning at 14%. 

Cash is also where changes in rates, trade policy and customer behaviour can become operational very quickly. Funding costs can move, supplier outflows can change, and collections can weaken before a long-range plan is formally recast. The survey findings place resilience much closer to the timing of receipts and payments, working capital and available liquidity.

Protiviti’s recommendations bring the cash agenda down to specific working capital decisions. Finance teams are being encouraged to accelerate receivables, reassess payment terms, review inventory levels, identify higher-risk customers and suppliers and tighten collections discipline. Stronger cash flow forecasting, working capital dashboards and liquidity reporting can meanwhile give management a clearer view of cash positions and funding needs before month-end. Those cash implications increasingly need to feed into sourcing, pricing, inventory, capital allocation and investment decisions across the business.

A forecast built around fixed assumptions can age quickly when tariffs, interest rates, supplier costs, customer payment behaviour, and demand can all move. Yet AI use remains much deeper in forecasting and risk than in the scenario work that connects those changes to liquidity. Among finance organisations already employing AI, 76% use it for financial forecasting and 67% for risk assessment, compared with 44% for scenario planning and 40% for cash flow management. Protiviti sees scope to use rolling scenarios to model changes in rates, tariffs, trade policy, supplier costs and customer behaviour, then stress-test the assumptions underpinning financial plans before those shifts reach the cash position. 

Rolling scenarios can also make the link between treasury and the rest of the business more explicit. A tariff change may begin in procurement, a customer slowdown in sales and a funding requirement in treasury, but each can ultimately alter the same cash forecast. The survey report recommends bringing operations, tax, deal, and product leaders into scenario planning so finance has a wider view of the assumptions underpinning its numbers. 

AI adoption outruns strategy

AI use has broadened further, from 72% of finance organisations last year to 77% in 2026. The bigger gap is between adoption and strategy: only 14% use it under a defined plan, while another 63% already use AI without one. 

Only 1% of finance organisations place themselves at the transformation stage, where AI is driving significant change across the function. A further 11% are at optimisation, 42% describe themselves as defined, 38% remain in experimentation and 8% are still at the initial stage. 

Public companies are further ahead: 62% rate their maturity at either defined or optimisation, compared with 42% of privately held companies. 

Usage is concentrated in familiar finance activities. Among organisations employing AI, 76% use it for financial forecasting, up from 58% last year. Risk assessment and management reaches 67%, process automation 56%, compliance and regulatory reporting 50%, scenario planning 44%, expense management 42% and cash flow management 40%. 

Some of those year-on-year movements are revealing. Scenario planning rose from 38% to 44% and cash flow management from 33% to 40%, while process automation slipped from 66% to 56%. Finance appears to be testing AI more often in analytical and decision-support work, as well as routine process execution. 

Governance becomes more important as use spreads across core processes. Forecasting, compliance, risk assessment and cash management all rely on controlled data and repeatable decisions. Protiviti’s recommendations therefore put data governance, access controls, model oversight and change management alongside the technology itself. Scaling AI across finance means deciding who owns the output, where human review remains necessary and which measures determine whether a use case should expand. 

Several of those use cases sit close to the heart of liquidity and risk management. Forecasting can pull forward signals from sales, costs and external data, while cash flow models can be refreshed more frequently and exposures tested against changing assumptions. Scenario planning and cash flow management still sit well below forecasting in adoption, leaving room for finance functions to connect these capabilities more closely.

Forecasting becomes the proving ground

Forecasting is becoming finance’s clearest test of what AI can actually deliver. Among publicly held companies, 81% are employing AI for financial forecasting. AI is arriving in forecasting just as the assumptions beneath those forecasts are becoming harder to hold. New products, changing markets, policy shifts and volatile input costs can invalidate a plan quickly. The appeal is monitoring a wider set of indicators and spotting changes earlier than a traditional periodic forecast cycle allows. 

Volatility findings make that especially relevant. A finance team that refreshes cash and earnings assumptions monthly can still be reacting to information that moved days or weeks earlier. More frequent scenario analysis can give treasury earlier visibility of funding needs, payment delays, liquidity risks or working capital pressure.

The report recommends keeping people firmly in the loop, with human review, model oversight and data quality becoming more important as finance moves towards semi-autonomous workflows. Faster analysis does little for treasury if the numbers feeding reporting and liquidity decisions cannot be trusted. 

ROI remains the uncomfortable question

Widespread deployment has moved faster than finance’s ability to prove AI’s value. Only 35% of finance organisations rate themselves at least moderately effective at measuring AI ROI. The comparable figure for broader business transformation is 45%. 

Measurement is inconsistent across technology spending more broadly. Some 64% have a formal methodology for calculating and measuring returns on technology investments, leaving more than a third without one. Public companies are further advanced here too, with 75% reporting a formal methodology. 

As a result, capital allocation becomes harder. AI may touch forecasting, reporting, compliance and automation, but enthusiasm alone cannot show whether a deployment saves enough time, improves accuracy or cuts enough errors to justify the investment.

The report recommends setting KPIs before scaling projects, including time and cost savings, forecast accuracy, error rates, cycle times, and improvements in decision quality. Those measures are familiar territory for finance and provide a firmer basis for comparing AI spending with ERP work, automation projects or broader transformation. 

A cash forecasting tool should show whether forecast variance improves. Automated reconciliation should reduce exceptions or processing time. Liquidity analytics should provide earlier or more reliable signals for funding decisions. Making those outcomes measurable gives finance something more concrete than adoption rates.

Proven automation still delivers the savings

Measured cost gains are still coming mainly from established technologies. Automation and robotic process automation lead by a wide margin, with 78% of organisations reporting meaningful progress from these tools over the past year, up from 59% in 2025. Technology rationalisation follows at 65%, while 60% cite cloud-based systems. AI and machine learning have climbed quickly to 44% from 28%, but still trail the other three. 

Headcount reduction, by comparison, is cited by only 23%, down from 36% last year. Offshoring, outsourcing and business process outsourcing reach 33%. 

Asked which technology approaches deliver the most valuable cost and efficiency benefits, advanced analytics ranks first for 25% of respondents. Process transformation follows at 22%, automation and RPA at 21%, AI and machine learning at 15% and ERP enablement, technology modernisation and reconciliation tools at 10%. 

Reported savings show a notable shift. Headcount reduction has fallen from 36% to 23%, while automation, technology rationalisation and cloud usage have all strengthened. In this year’s responses, process and technology change, rather than workforce cuts, is driving more of the efficiency agenda.

Clean data, integrated systems, automated routine work, and usable analytics still deliver much of the measurable value. They also determine how far AI can scale. Protiviti identifies stronger data quality, governance, integration and access as prerequisites for scalable automation and AI. 

Automating reconciliation or improving system integration can create value before AI enters the process. Once those foundations are working, AI has cleaner data and more dependable workflows to draw on. A cash forecast cannot improve if underlying data arrive late, entity structures are inconsistent or ERP feeds still require manual repair. 

Security keeps its grip on the agenda

Data security and privacy remains finance’s number one priority for the third consecutive year, scoring 7.6 out of 10. Financial planning, profitability analysis and reporting ranks second at 7.3, followed by enhanced data analytics at 7.0. 

AI has made the biggest move. It rises from 13th place last year to sixth across all respondents, with a score of 6.9. Among CFOs and vice-presidents of finance, it ranks fourth, up from 14th. 

AI, cloud applications, advanced analytics and automation are putting more finance data into more systems. That makes access controls, privacy safeguards, data governance and model oversight part of the transformation itself. 

Cash management and forecasting depend on the same finance data that feeds analytics and automation. Greater automation can improve visibility, but it also increases the importance of knowing which systems can access sensitive data and how outputs are validated before they influence funding, investment or other finance decisions.

Cash becomes the test of transformation

Protiviti’s findings leave finance leaders with a broader challenge than deciding where to deploy AI. Technology initiatives are arriving at the same time as cash management, forecasting, cost control and resilience demand more attention. 

Cleaner data strengthens both analytics and forecasting, while automation improves the information feeding cash management. ROI discipline then determines which tools deserve further capital, and governance determines how far they can safely scale.

Ultimately, the technology agenda lands in treasury as a visibility problem. Cash has to be visible before it can be moved; forecast changes need to surface before funding decisions are locked in, and weakening assumptions need to be identified while there is still time to respond.

Month-end is a poor place to discover that the assumptions beneath a cash position have already moved. The advantage lies in seeing that shift early enough for treasury to act.