Companies are spending hundreds of billions on AI this year. Most cannot tell you what it is returning.

Worldwide AI software spending is forecast at roughly $453 billion in 2026, up from $283 billion in 2025, according to Gartner. 88% of executives report their AI-related budgets will increase in the next 12 months, according to PwC's AI Agent Survey. Only 4% of CFOs maintain a conservative AI strategy today, down from 70% five years ago.

The money is flowing. The conviction is real. The question that most organizations cannot answer is the one that matters most: what is all of this spending actually producing?

PwC's 2026 Global CEO Survey delivers the uncomfortable answer: 56% of enterprises have captured neither revenue gains nor cost savings from their AI investments.

More than half.

The Scale of the Spend

The commitment to AI is no longer selective. It is near-universal.

Hundreds of billions in AI software spend. 88% of executives increasing budgets. The CFO holdout population has collapsed from 70% to 4% in five years. There is no meaningful executive resistance left to AI investment as a category.

This is not a story about early adopters taking risks. This is a story about the entire enterprise market moving in the same direction, at the same time, with historically large budget allocations.

The question is whether that spend is generating returns proportional to the commitment. For most organizations, the answer is no.

The Value Capture Failure

Deloitte surveyed 3,235 director-to-C-suite leaders across 24 countries. The findings create a picture of widespread deployment with limited impact.

Deployment is now near-universal, but value capture is not. PwC's 2026 Global CEO Survey found that 56% of enterprises have captured neither revenue gains nor cost savings from AI.

The gap between AI budget growth and AI value capture is the largest disconnect in enterprise strategy right now. Companies are doubling down on investments that more than half of them cannot connect to a financial outcome.

What Separates the Organizations Seeing Returns

The organizations capturing real value from AI share patterns that are observable and replicable. None of them are about choosing the right vendor.

Measurement discipline before deployment. The organizations seeing returns defined what success looks like before they deployed. They set specific financial metrics tied to business outcomes, not AI activity metrics. "Reduce customer support cost per ticket by 22%" is a success metric. "Deploy AI across 4 departments" is an activity metric. The winners optimized for the first. The strugglers tracked the second.

Defined success metrics tied to business outcomes. Revenue impact, cost reduction, cycle time compression, customer retention improvement. The organizations seeing returns measured AI against the same metrics they use to evaluate any other strategic investment. They did not create a separate measurement framework for AI. They held AI to the same standard as everything else.

Workflow redesign over tool layering. This pattern appears in every data set on enterprise AI success. The organizations seeing returns redesigned how work gets done before deploying AI into the workflow. The strugglers added AI to existing processes and hoped for improvement. Deloitte's finding that 84% of companies have not redesigned jobs around AI maps directly to the value capture failure.

Leadership that owns the ROI question. In the organizations seeing returns, the CEO or CFO owns the AI ROI conversation. It is not delegated to a Chief AI Officer or a VP of Innovation. The person accountable for company financial performance is also accountable for AI financial performance. That alignment changes the decisions that get made about where, how, and whether AI gets deployed.

The Question for 2026

The spending will continue. No major enterprise is reducing its AI budget this year. The 88% trajectory is locked in.

The question is whether the spending will start producing proportional returns, or whether the gap between investment and impact will widen further.

For business leaders, the honest assessment starts with a simple exercise. Take your total AI spend for the last twelve months. Then list every measurable financial outcome it produced. Revenue generated. Costs reduced. Efficiency gains that translated to margin improvement.

If the list is short, the problem is not the technology. The technology works. The organizations seeing returns proved that. The problem is how the technology was deployed, measured, and managed.

Hundreds of billions of dollars is a large number. The organizations that treat it as a strategic investment with specific return expectations will be the ones that justify it. The ones that treat it as a cost of keeping up will continue to spend without knowing what it produces.

The difference between those two approaches is not budget. It is leadership.

Related: how Jackson runs AI agents as an executive team and work with Jackson on AI systems.