85% of Financial Firms Pour More Cash Into AI Budgets

85% of financial services and insurance firms plan to raise AI budgets over the next 12 months, a signal that enterprise AI has moved from test bench to capital plan. The finding comes from a PYMNTS Intelligence report based on a March survey of 60 senior technology executives at U.S. enterprises with at least $1 billion in annual revenue, according to PYMNTS.

The sharper read: financial firms AI budget increases are not about novelty anymore. They point to a race for operating leverage, faster risk decisions, better productivity, and more defensible competitive positioning. Finance is pulling ahead because its best AI use cases sit inside structured, auditable workflows where outcomes can be measured.

“Enterprise artificial intelligence (AI) is no longer just a technology project. It is becoming a line item with momentum.”

Financial firms AI budget increases are now a capital-allocation signal

The 85% figure matters because it shows AI moving into planned spending cycles. In finance, that usually means the conversation has shifted from “can this work?” to “where does this fit in the operating model?”

The report says financial services and insurance firms are funding AI for three reasons: productivity, competitive positioning, and risk reduction. Two of those are especially measurable. 65% cite productivity and efficiency gains as a reason for AI investment, while another 65% cite strategic or competitive positioning.

That combination explains why finance is ahead of the surveyed sectors. Banks, insurers, payments firms, and fintechs already run through controlled processes: accounting close, credit assessment, risk scoring, customer checks, forecasting, and reporting. Those workflows give AI something to work with.

XOOMAR analysis: the firms most likely to gain are not the ones with the most visible AI branding. They’re the ones that can place models inside daily work without weakening controls. In finance, trust is not a marketing layer. It’s the product.

The survey shows finance ahead, but not alone

The PYMNTS report compares financial services and insurance, healthcare, and media and advertising. All three sectors plan to spend more on AI, but the spending logic differs.

Financial services also leads on task adoption. A related PYMNTS summary says the sector reached high adoption on 27 of 75 AI-supported tasks, compared with 16 in media and advertising and 10 in healthcare. That means finance alone reached more high-adoption tasks than the other two sectors combined.

The strongest finance use cases are not experimental consumer tools. They are structured finance and risk functions:

• Revenue recognition and accounting close: 65% adoption in financial services and insurance.
• Credit risk assessment and scoring: 60% adoption.
• Sales forecasting and pipeline optimization: 60% adoption.
• Churn prediction and retention targeting: 30% adoption.
• KYC, KYB, and identity verification: 20% adoption.
• A/B testing: 10% adoption.

That pattern is the story. AI has gained the most ground where rules, historical data, and validation paths already exist. It lags where customer behavior, product design, and identity workflows introduce messier data and higher execution risk.

Measurable workflows are winning the first AI spending round

Finance firms are spending on AI where the payoff can be judged against existing benchmarks. If a model helps close the books faster, improves credit scoring, sharpens forecasting, or reduces operational drag, executives have something concrete to assess.

That is different from media and advertising, where 80% expect to raise AI budgets but only 25% point to monetary ROI and financial metrics, the lowest share in the survey. Media firms are moving quickly, but the report suggests their financial case is less mature.

Healthcare shows another pattern. 60% of healthcare and medical firms plan to spend more, and 60% cite pilot funding with no formal ROI requirement. The report frames that as a sector still testing what works, especially under operational pressure and fragmented systems.

XOOMAR analysis: financial firms have a cleaner first wave because many AI projects can be attached to existing control points. The next wave will be harder. Moving from accounting close and credit scoring into customer-facing decisioning requires stronger data plumbing, clearer accountability, and tighter model oversight.

That control theme also runs through broader fintech coverage. Readers tracking governance in finance may want to revisit Legal Trap Falls as SEC Lets UBS Crisis Plan Move Forward, while the AI banking angle has also surfaced in deal narratives such as Axos Arc Acquisition Pulls AI Banking Into M&A Fight.

Data fragmentation is the brake on the AI race

The report is clear that money alone will not decide who wins. 30% of financial services executives cite data quality and fragmentation as the top barrier to further AI deployment.

That obstacle matters because finance has decades of transaction histories, risk records, customer files, payment data, and compliance documentation. But having data is not the same as having usable data. Fragmented systems make it harder to train, test, monitor, and explain AI outputs across the business.

Healthcare faces a double constraint. 30% point to system integration, and another 30% point to data quality. Media and advertising face a broader set of obstacles, including skills gaps, governance, and leadership alignment.

For financial firms, the message is sharper. They appear to have stronger budget momentum and broader adoption, but their next constraint is internal architecture. If data remains fragmented, AI stays trapped in the parts of the company where the information is already clean enough to use.

CFOs, CIOs, risk teams, and customers will judge AI differently

The PYMNTS survey measures technology executives, not every stakeholder inside the firm. Still, the data points to a real internal tension.

A CIO can view AI as modernization. A CFO will ask whether the spend produces efficiency, revenue support, or margin improvement. A risk leader will focus on data quality, model reliability, and governance. Business heads will want faster decisions and better tools.

Customers are not directly surveyed in the report, so their reaction is an open question. But the source-supported constraint is obvious: finance firms are moving fastest in structured functions and slower in customer-facing areas such as churn prediction, identity verification, and A/B testing. That gap suggests firms are more comfortable deploying AI where controls are already mature.

The report also undercuts the most extreme automation narrative. Across sectors, 80% to 85% of executives say AI will augment decision-making over the next five years. They do not expect a full handoff to machines.

That matters for budget discipline. If AI is mainly funding better tools for people, then adoption metrics should focus on how work changes: faster close cycles, better forecasts, stronger risk scoring, fewer manual steps, and clearer escalation paths.

Budget growth will expose disciplined operators and budget tourists

The financial firms AI budget increases signal is real momentum, but it is not proof of durable advantage. Spending can rise because projects are working. It can also rise because firms are buying tools faster than they retire old workflows.

For operators, the practical test is whether AI is tied to specific tasks with measurable outcomes. The PYMNTS data suggests finance is already doing this in revenue recognition, accounting close, credit risk assessment, and forecasting. Those are the places where AI can be judged against known processes.

For vendors, the opportunity is strongest where financial firms have named pain: data fragmentation, integration, governance, and task-specific deployment. But the report does not prove which vendor categories will benefit most. That remains outside the data.

For investors, AI budget growth is a signal to interrogate, not a metric to accept at face value. The better questions are simple:

• Adoption: Is AI used in core workflows or isolated pilots?
• Measurement: Are productivity and efficiency gains tracked?
• Controls: Can the firm monitor model outputs and data quality?
• Expansion: Can AI move beyond back-office functions without adding new risk?

Over the next 12 months, the evidence that would confirm the thesis is clear: broader task adoption, fewer data fragmentation barriers, and more AI use cases tied to hard operating metrics. The evidence that would weaken it is just as clear: bigger budgets with no measurable productivity gains, no movement in customer-facing adoption, and persistent data quality problems.

The winning financial firms will make AI less flashy and more useful. Controlled, measurable, deeply embedded. That’s where the money is starting to go.

Disclaimer: This XOOMAR analysis is for informational and educational purposes only. It is not financial, investment, legal, tax, or professional advice. It does not provide buy, sell, hold, price-target, portfolio, or personalized recommendations. Verify information independently and consult qualified professionals before making decisions.

The Bottom Line

• AI is moving from experimentation into formal capital planning at major financial firms.
• Banks, insurers, payments firms, and fintechs are targeting AI where workflows are structured and measurable.
• The spending shift could widen the gap between firms that operationalize AI effectively and those that do not.