Ahmed Atia explains why AI in finance requires consistent data, shared forecasting assumptions and human accountability across GCC businesses
Giving artificial intelligence a larger role in financial forecasting, reporting and consolidation gives it greater influence over management decisions. For CFOs and finance leaders across the GCC, successful AI adoption requires more than new technology. It requires consistent data, shared business assumptions and clear human accountability.
Forecasts shape investment, and recommendations influence which businesses receive resources. Yet businesses risk placing confidence in AI-generated conclusions before resolving disagreements about data, assumptions and responsibility. As AI in finance extends to more activities, those disagreements can spread across processes and become harder to detect.
McKinsey’s 2025 global survey of 102 CFOs across industries found that 44 per cent reported using generative AI for more than five use cases, compared with 7 per cent in the previous year’s survey. That expansion raises an organisational question: can individually credible applications produce a coherent view of the business? A forecast and a performance report may each appear sound while relying on different definitions or assumptions.
Management must resolve those differences if AI-supported analysis is to provide a dependable basis for decisions.
Data consistency in financial forecasting and consolidation
For GCC businesses operating across sectors and markets, meaningful financial consolidation depends on information that management can compare and interpret. Bringing figures together does not mean they describe the business on consistent terms.
Consider two subsidiaries that classify overdue receivables differently, while a third updates its collection assumptions less frequently. An AI model could produce a convincing cash forecast from those inputs, concealing differences that materially affect the group’s expected cash position.
A shared platform makes information accessible, but data quality depends on agreed definitions, reporting periods and exceptions.
Technology teams can implement those decisions, but business leaders who understand both group requirements and local operations must settle them.
Consistency must still accommodate legitimate commercial differences, since a retailer and a contractor cannot forecast receipts identically. Those differences should remain visible in management reports, so leadership can assess uncertainty rather than mistake a uniform presentation for comparable information.
Why finance and operating teams must share forecast assumptions
Financial systems record transactions, but knowledge about what happens next often sits with the people managing customers, contracts and delivery. The growing use of AI in financial forecasting makes that operating knowledge more consequential.
A delayed order could reflect a scheduling issue or a weakening customer relationship. Identifying the movement does not establish its meaning, which requires input from someone close to the account.
When operating teams are consulted only after a forecast is complete, their contribution becomes a negotiation over the numbers, arriving when finance is already defending the result.
AI could reinforce this pattern, with commercial teams generating scenarios that support ambitious projections while finance produces analyses that favour caution.
Shared ownership of assumptions offers a stronger basis for decisions. Differences of judgement can remain, provided management understands which assumptions it has accepted and where uncertainty persists.
As an illustration, suppose a business unit’s largest customer is now expected to pay later than forecast. The business owner explains the commercial reasons for the change, finance challenges its implications for cash flow and funding, and the analysis is refreshed using the approved assumptions. Management then decides whether to revisit payment terms, defer expenditure or secure additional liquidity.
AI governance requires clear human accountability
Responsibility can blur when several teams contribute to an AI-supported conclusion. Those supplying information may assume finance will validate the result, while finance assumes technology colleagues have confirmed the system can be trusted.
Yet a technically reliable system can still produce a misleading conclusion from incomplete information or inappropriate assumptions.
Effective AI governance assigns each responsibility explicitly. Operating leaders should own the commercial assumptions, finance should validate their financial implications and technology teams should maintain system reliability. A named decision owner should then approve the resulting action, so accountability rests with an identifiable person rather than a process.
Human review provides assurance only when reviewers can examine the evidence, challenge the interpretation and stop an unsupported conclusion from informing a decision. Organisations must therefore preserve the experience that allows employees to recognise when a plausible answer conflicts with business reality.
Finance transformation requires organisational change
For GCC businesses, the challenge of finance transformation is maintaining a dependable view of performance as operations grow more complex. AI in finance can support that ambition, provided each leader takes clear responsibility for their part of the information and judgement behind its outputs.
The test is whether management can use that analysis to make decisions it can explain and defend. Organisational change gives AI a credible role in those choices, while keeping responsibility firmly with the people making them.
This opinion piece is authored by Ahmed Atia, Director, Corporate Finance and Business Control, Raya Holding for Financial Investments.
Source: Tahawul Tech
