The rapid evolution of AI economics has caught many organizations off-guard.

More than half of surveyed executives are now concerned about AI costs, up from 45% 12 months earlier. Nearly three in five executives said their organization’s AI expenditure was significantly higher than expected or budgeted in the past 12 months. Among those reporting an overrun, expenditure was 24% above expectations or budget on average.

The challenge extends beyond controlling expenditure. Many organizations lack a complete view of the direct and indirect costs associated with AI, making it difficult to forecast spending, demonstrate returns, and identify which investments create measurable value.

Our research snapshot, The cost of intelligence examines examines the changing economics of enterprise AI and how organizations can connect spending more closely to outcomes. The findings are based on a September 2026 survey of 500 senior executives.

Our findings at a glance

Cost overruns are becoming commonplace
Nearly three in five executives reported significantly higher-than-expected or budgeted AI expenditure in the past 12 months.

Most organizations lack an enterprise-wide view of AI costs
60% of organizations still lack consistent, enterprise-wide visibility into AI costs

A significant share of expenditure cannot be tied to verified outcomes
Among organizations able to estimate expenditure coverage, 61% of AI spending can be systematically tracked, but only 45% can be connected to verified business value.

Organizations are tightening controls rather than rethinking AI delivery
Among organizations reporting cost overruns, 53% responded by controlling expenditure, while 20% reconsidered the tools or processes involved.

AI investment decisions are becoming more cross-functional
CFOs now help approve major AI investments at 52% of organizations, up from 42% a year ago. Involvement from business-unit leaders and chief AI or data officers has also increased.

Improving the economics of AI at scale

Organizations can improve AI economics by making the full cost and value of AI outcomes visible through AI FinOps and outcome-based metrics. Clear accountability and spending guardrails can strengthen governance, while smarter model selection and workflow design can improve cost, quality, and reliability. The Snapshot also recommends using AI Yield to guide investment decisions and fostering a culture of responsible experimentation that rewards measurable value over AI activity.

Read the full AI Economics Snapshot

Explore why AI costs are exceeding expectations, where visibility breaks down, and how organizations can connect AI expenditure more closely to verified business value.