AI costs are now a concern for 53% of executives according to a new ‘snapshot’ report by the Capgemini Research Institute, based on a September 2026 survey of 500 CXO-level senior executives from large organisations.

Beyond direct cost such as tokens, executives lack visibility into the full cost of safe AI deployment, including human oversight, rework and governance.

To address this, businesses are both tightening cost controls and exploring multi-model strategies.

As organisations scale AI and prepare for more agentic use, the research highlights a need for greater focus on the business value generated by AI transformation.

A lack of cost visibility across the enterprise risks serious financial impact

In the past 12 months, 59% of executives say their organisation’s AI-related expenditure was significantly higher than expected or budgeted. Among the organisations affected, expenditure was 24% above expectations or budget on average, and around one in five organisations (21%) overshot their AI budget by more than 30%.

Concern about AI costs has risen over the past year, and only 56% are fully confident they can forecast how AI costs will change as adoption scales.

The report points to a lack of cost visibility across the enterprise. Half of executives say AI spending is approved without adequate visibility into the full cost.

Organisations also lack confidence in accurately estimating the full cost of their existing AI initiatives, particularly indirect costs such as human oversight, rework and governance.

The financial impact can be serious for those affected. Two thirds of organisations that overshot their AI budget faced negative financial consequences, such as lower free cash flow or a reduction in discretionary spending. In that context, CFOs are now more involved in approving major AI investments.

“As AI moves deeper into business workflows, organizations need to stop treating its cost as a technology or procurement issue and start managing AI as a portfolio of economic outcomes,” said Volker Darius, CEO of Capgemini Invent.

“Cost discipline alone will not create a profitable AI estate, just as value creation without cost control will not be sustainable. Organizations need to measure cost, quality and business value together, use the most appropriate model for each task, and account for the human oversight required to deliver outcomes safely. By continuously measuring AI Yield, organizations can make more informed capital-allocation decisions and scale AI with confidence.”

Multi-model strategies lead AI portfolio plans

In response to unexpected AI spend, over half (53%) of organisations increased expenditure control, either by tightening budgets or controls for the affected initiative, introducing organisation-wide AI spending controls, or reducing the initiative’s scale or scope.

Only one in five rethought tools or processes, including which activities should be AI-led rather than human-led, or changed the AI solution, models, configuration or model routing.

Beyond these immediate responses, more than three in five executives expect their organisation to adopt a multi-model strategy over the next 12 months, dynamically selecting models based on cost, capability, latency, risk, and other business requirements.

Executives expect greater use of lower-cost proprietary models where they are sufficient, as well as small or specialised models for specific tasks. Nearly three in five organisations also plan to expand their use of open-weight models.

Recommended reading

• UK AI Funding Reaches Record £4.56BN in “Landmark” Quarter
• Gartner Forecasts AI Will Drive a Net Increase of 1.7 Million Jobs by 2030
• UK Budget Unveils Major AI Investment Package
• Beyond The Hype: Turning AI Investment Into Business Value

AI Yield brings successful business outcomes into focus

Although two thirds of executives say they are confident they can verify the business value generated by AI-assisted outcomes, most organisations cannot directly match AI spending with successful business outcomes – either because they do not formally measure them, because they track costs or usage but not successful business outcomes or because they track them separately.

Demonstrating sufficient value to secure approval and funding for AI investments remains particularly challenging, with 57% citing it as their biggest obstacle.

Only one in five organisations calculates AI Yield, a metric that captures the verified value of an AI-assisted outcome after accounting for its full safe-delivery cost.

The business value most often cited is employee productivity or released capacity followed by operating-cost reduction and faster processes or cycle times. Less than a third (32%) cite revenue growth, including growth from new products, services or business models.

The metrics used to manage AI economics are therefore expected to change. Today, 60% primarily use total AI expenditure against budget, but only 44% expect to prioritise it in the next 6-12 months.

First-pass success rate is set to become the top-priority metric, and cost per completed task or workflow is also expected to be prioritised. These measures put successful outcomes, rather than AI consumption alone, at the center of AI economics.