AI adoption in finance doubles, but assurance readiness determines who wins, KPMG finds

23 June 2026 — KPMG has launched 2026 Global AI in Finance: The Decision Advantage survey , drawing insights from 1,013 senior finance leaders across 20 countries and 13 sectors, examining how AI is being deployed, measured and governed within the finance function. The findings signal a shift: AI has moved rapidly from experimentation to impact. Active use of AI in finance has more than doubled since 2024, rising from 30 percent to 75 percent, and nearly three-quarters of leaders (71 percent) say it is meeting or exceeding ROI expectations, with the strongest returns appearing in decision-making quality, forecast accuracy and responsiveness.

AI is pushing finance beyond basic process automation and into areas where judgement matters most such as forecasting, planning, and risk assessment, where finance has traditionally struggled. For example, 70 percent of organizations report improved decision-making quality, faster decision-making (71 percent), and stronger forecast accuracy over the past year (64 percent).

However, performance varies sharply by sector. In banking, 71 percent of leaders report moderate or significant improvements in forecast accuracy, compared with just 44 percent in healthcare—a 27-point gap driven by differences in data foundations. Banking benefits from structured data that gives AI a foundation. By contrast, sectors such as healthcare face fragmented data sources that limit what AI can do. Ultimately, stronger data foundations are what enable judgment-driven AI use cases to scale and perform.

As AI becomes more embedded in finance processes, organizations with strong controls, governance, and auditability are seeing a clear performance advantage. The report suggests assurance-ready organizations, those that are able to produce audit evidence and explain it, are delivering stronger outcomes, reporting three to six times higher rates of improvement in error reduction than their peers (33 percent versus 6 percent) and are more confident in their ability to scale AI (42 percent versus 14 percent). When finance teams can clearly demonstrate how AI decisions are generated, trust shifts from assumption to proof, enabling more consistent performance and more confident adoption at scale.

At the same time, the report suggests many organizations are still building this capacity. Fewer than half of organizations (42 percent) are fully assurance-ready for AI-enabled finance processes, and only 29 percent track where AI adoption fails, creating partial visibility that shows what AI delivers but not where it breaks down, such as how outcomes are produced or where risks like errors or bias may emerge. As regulators and external auditors increasingly require proof that AI-enabled finance processes are controlled and explainable, assurance-readiness becomes even more essential for organizations to sustain performance and maintain trust with stakeholders.

The biggest barrier to AI success isn’t technology, it’s data. The survey shows that 36 percent of organizations cite data quality as both their biggest barrier and opportunity, pointing to improvements in data integration and system interoperability as key to unlocking more value from AI in finance. In response, most organizations are focused on upskilling their existing finance and internal audit teams (38 percent), while fewer are choosing to hire for new skills (28 percent). The organizations pulling ahead to close this data quality gap are choosing to do both: upskilling existing teams while also bringing in people who can assess data quality and interpret outputs.