The question used to be whether AI would deliver real business value. According to NVIDIA’s 2026 State of AI report, that debate is over.

The survey polled 3,200 organisations across industries and regions, and the results are unambiguous: 88% of respondents report AI-driven revenue increases, with 30% saying their gains exceeded 10%. On the cost side, 87% achieved measurable savings, with the retail and consumer packaged goods sector leading — 37% of those businesses cut costs by more than 10%.

This is not a story about pilots and proofs of concept. Nearly two thirds of organisations (64%) are now actively deploying AI in operations, up significantly from the exploratory phases that defined 2024 and early 2025. Only 3% of respondents said they had written off AI entirely.

What the Numbers Actually Say

The headline figures matter, but the data underneath them tells a more nuanced story about where value is being captured and where it is being left on the table.

Data analytics is the top AI workload at 62% of respondents, with generative AI close behind at 61%. That convergence matters: the companies pulling ahead are not treating data and AI as separate initiatives. They are using AI to do something with their data, not just store it.

Budget commitment is also strong. 86% of respondents said their AI budget will increase in 2026, and nearly 40% expect increases of 10% or more. That is not cautious experimentation — that is conviction.

The spending priorities reflect a shift in maturity. The top priority for 42% of respondents is optimising existing AI workflows and production cycles. Finding new use cases comes second at 31%. The industry has moved past the “what can AI do” phase into the harder and more valuable “how do we do it well” phase.

The Gaps That Still Exist

Not every organisation is winning equally. The NVIDIA data surfaces two persistent problems that are holding businesses back.

Data quality is the top challenge, cited by 48% of respondents. That is not surprising. AI can only be as good as the data it learns from, and most businesses have spent years accumulating data without treating it as a strategic asset. Messy, inconsistent, poorly governed data is the hidden tax on every AI initiative.

The AI expertise gap is close behind at 38%. Nearly a third of respondents are also still struggling to quantify ROI, which creates a budget approval problem even when the work itself is delivering results.

North America leads global adoption at 70% of organisations actively using AI. But even there, the internal skills shortage is a limiting factor. Having access to AI tools and knowing how to extract value from them are two different things.

What This Means for Business

If your business is in the 12% not yet seeing revenue gains from AI, the data suggests three likely explanations: your data foundations are weak, your team lacks the skills to implement well, or you have not moved beyond experimentation into real operational deployment.

The companies in the 30% who are seeing 10%+ revenue gains are not smarter or better resourced. They have invested in the foundations: clean data, trained teams, and processes designed around AI augmentation rather than bolt-on tools.

The budget trend reinforces urgency. When 86% of your competitors are increasing AI spend, staying flat is not a neutral choice. The gap between AI leaders and laggards compounds quickly.

For most business leaders, the path forward looks like this: fix your data quality, build genuine AI literacy across your team, and focus your AI investments on the workflows that touch revenue or costs most directly. These are not glamorous steps, but they are the ones that separate the 88% from everyone else.

Enterprise DNA exists at exactly this intersection. Our learning platform gives data and AI skills to the people inside your business who need them. Our Omni services deploy the agents and systems that put those skills to work. The combination is what gets you into the group seeing real returns.

The proof is in the data. Literally.

---

If this is the kind of problem agents can help with, the free Working With Claude field guide is the practical next step. Thirty-two pages, no fluff. Get the free guide.