Almost three-quarters of large businesses are using AI regularly. Yet only one in ten say it’s actually core to how they operate.

That’s the central finding of Publicis Sapient’s 2026 Global Enterprise AI Report, released on June 17 at VivaTech in Paris. The firm surveyed 1,550 AI decision-makers across six major markets, and the results tell a clear story: widespread AI adoption is masking a much smaller number of genuine AI-native operations.

For business leaders trying to figure out where their company stands in the AI transition, this report is worth understanding.

The Numbers That Matter

The headline stat is stark: 73% of respondents say AI is used regularly or across most of their business processes. But drop down one level and the picture changes. Only 10% describe AI as core to how their business actually operates.

That’s not a rounding error. It’s a structural gap.

A few more data points from the report:

• 47% believe AI is already fully capable of meeting today’s business needs
• 42% agree AI is capable, but say their organizations simply are not built to capture that value
• Only 38% say AI is fundamentally changing how their business operates
• 22% name organizational design as the primary constraint to AI success, not the technology itself

The UK stands out as the most transformed market in the survey, with 51% of UK respondents saying AI is fundamentally changing their operations and 60% reporting AI is highly or fully embedded into their workflows.

At an industry level, telecoms led on agentic AI deployment at 48%, with retail and consumer goods close behind at 47%.

What This Gap Actually Means

The data is telling us something that anyone who has tried to implement AI inside a real business already knows: deploying a tool and reorganizing around it are two completely different things.

Most businesses are in the first category. They’ve subscribed to Copilot, added an AI assistant to their help desk, run a few pilots, and checked the “AI in use” box. That’s not transformation. That’s experimentation that happened to stick.

The 10% who say AI is core to operations have done something different. They have restructured workflows, changed how decisions get made, trained their teams differently, and rebuilt processes around AI outputs rather than human ones.

The 42% who say “the tech is ready but our org isn’t” is the most honest number in the report. They know what needs to happen. They just haven’t done it yet.

And the finding that 22% name organizational design as the primary constraint is particularly significant. The barrier is not model quality, API access, or compute cost. It is the human and structural layer that makes AI work at scale.

What This Means for Business

If your business is in the 73% that uses AI regularly, the relevant question is not “are we using AI?” It is “is AI changing how we operate, or are we just layering it on top of old workflows?”

There is a meaningful difference between:

• Using ChatGPT to draft emails versus redesigning your client communication workflow around AI
• Running an AI proof of concept versus measuring AI-driven productivity gains at the team level
• Having an AI tool available versus having AI outputs feed directly into business decisions

The report suggests most businesses are doing the first option in each of those pairs. The gap between adoption and integration is where the real competitive advantage is being built right now.

The businesses that bridge this gap will not just be more efficient. They will operate in fundamentally different ways to their competitors, with faster decision cycles, lower overhead per unit of output, and teams focused on higher-value work.

The Organizational Design Problem

The 22% who identify organizational design as the core constraint are naming the real problem clearly. You cannot bolt AI onto an org chart built for a world without AI and expect it to become core to operations. The org chart itself has to change.

That might mean restructuring how teams collaborate, what decisions get made where, which roles exist versus which tasks get automated, and how performance gets measured. These are not technology problems. They are leadership and change management problems.

This is also why the gap between 73% and 10% is so hard to close. Most businesses are comfortable deploying AI tools. Far fewer are comfortable redesigning how their organization functions.

For small and mid-sized businesses, this is actually an opportunity. They have far less organizational inertia than large enterprises. A 50-person firm can rebuild workflows around AI in weeks in a way that would take a 50,000-person firm years. But only if leadership treats it as a transformation project, not a software rollout.

The Enterprise DNA Perspective

We talk to business owners every week who are somewhere in this 73%. They are using AI tools, some more than others, but few have asked the harder question: has anything about how we work actually changed?

The answer is usually no, or not enough. The tools are there. The workflows are not redesigned. The team is not yet thinking about their work differently. And so the productivity gains are incremental, not structural.

The Publicis Sapient findings match what we see directly. The technical capability is not the bottleneck. Getting an organization to restructure itself around AI is a leadership and operational challenge, and most businesses either do not know where to start or are uncomfortable with the disruption involved.

That is exactly the gap that work like Omni Advisory is designed to close. Not installing another AI tool, but helping leadership understand what it actually means to build a company where AI is core to operations, and then doing the work to get there.

The 10% who have made AI core to how they operate are not lucky. They made a deliberate decision, and then they executed on it.

The other 90% still have that choice to make.