AI Adoption Is Stalling Inside the Enterprise

Gallup’s latest data delivers an uncomfortable truth for enterprise leaders: while AI use is intensifying for some employees, enterprise adoption has stalled.

In Q4 2025, only 38% of U.S. employees said their organization has integrated AI tools to improve productivity, efficiency, or quality. That figure has barely moved quarter over quarter.

Since mid-2023, overall AI use in the workplace has more than doubled.

On the surface, that suggests acceleration. A deeper dive into the data, however, tells a more constrained story. The growth is coming from existing users, not from broad adoption across roles. Frequent and daily use are rising primarily among leaders, managers, and early adopters, while many individual contributors remain outside formal AI enablement.

This is how a stall hides in plain sight. The inside of the circle gets denser, but the circle itself does not expand.

Where the Stall Is Really Coming From

If widespread employee resistance were the primary constraint, AI use would likely be flat even among those with access. Gallup’s data shows the opposite. Usage continues to rise wherever AI tools are available and supported.

This suggests the stall is less about unwillingness and more about structure, clarity, and leadership direction.

Several leadership-level friction points continue to slow expansion:

• Data and security concerns that remain unresolved
• Governance models that lag real-world behavior
• A lack of role-specific, production-ready use cases
• AI framed as experimentation rather than operational work
• Managers unsure how to lead teams that use AI day to day

Until these issues are addressed from the top, adoption cannot scale beyond early adopters. AI use deepens where permission exists, but the organization as a whole remains constrained.

A Message to Employees: Don’t Wait to Be Ready

To be clear, none of this absolves employees of responsibility.

Even in organizations where formal AI integration lags, individuals still have agency. Employees who build AI literacy, experiment responsibly, and learn how to work alongside AI will be better positioned as adoption expands.

The risk of waiting is falling behind peers who are already learning new ways of working. The opportunity is to prepare now, even if the organization is not yet fully ready.

But individual initiative has limits.

Without leadership direction, AI use remains uneven, informal, and difficult to scale. Personal effort can accelerate readiness, but only leadership can turn that readiness into enterprise capability.

Leadership From the Top Is Non-Negotiable

AI reshapes how work gets done, how decisions are informed, and how risk is managed. These are not decisions that can be delegated downward and hoped into existence.

Only senior leadership can:

• Set acceptable risk tolerance
• Define where AI will and will not be used
• Fund integration instead of endless pilots
• Signal that AI belongs in core work, not side work

Absent that signal, AI adoption fragments. Shadow use hardens. The stall deepens.

The CIO’s Window

The current stall is not just a warning. It is an opening.

CIOs can translate executive intent into operational reality, turning isolated AI use into a shared enterprise capability. The organizations that move next will not be the ones with the most pilots, but the ones where leadership decides clearly and visibly that AI is part of how work gets done.

AI adoption is no longer a technology question. It is a leadership test.

When enterprise adoption stalls:

• Productivity gains remain uneven
• Capability gaps persist
• Shadow AI becomes permanent
• Momentum dissipates

The cost of waiting is not neutrality. It is drift.

Gallup’s data makes one thing clear. AI is becoming indispensable for those who are enabled to use it. What remains unresolved is whether CIOs will bring the rest of the organization along, and whether employees will be ready when they do.