Skip to content
Momentum Applied

AI & Operating Model · September 2026

AI is an operating decision, not a technology purchase.

The hard part of enterprise AI is not finding another model or platform. It is deciding where AI belongs in the operating system of the business — who owns it, what it can touch, how value is measured, and how people adopt the change.

The technology question arrives too early

Organizations often begin with tools: which platform, which model, which vendor. Those choices matter, but they become durable only after the business has defined the work that should change, the decisions AI may influence, the data it can use, and the accountability that remains human.

Start with the operating decision

For each AI opportunity, identify the process, decision, or handoff that needs to improve. Define the business owner, the current constraint, the expected change in behavior, and the measure that will show whether the intervention works. This makes AI part of operating design rather than a separate innovation track.

Governance should accelerate the right work

Useful governance is not a committee added after the pilot. It is a set of decision rights, entry criteria, risk thresholds, architecture patterns, and review cadence that lets low-risk work move quickly while escalating decisions that deserve executive attention.

Adoption is part of the architecture

If a new AI capability changes who makes a decision, how an exception is handled, or what a manager is expected to review, then training and communication are not enough. Roles, controls, metrics, and operating routines have to change with the technology.

Measure movement, not activity

Pilot counts, licenses, and model calls are activity measures. Executives need evidence that a process became faster, a risk became more visible, a decision improved, or capacity shifted to higher-value work. The measure should be defined before the pilot starts.

The practical test

Before funding another AI initiative, ask five questions: What business decision or process changes? Who owns the outcome? What data and controls are required? What human behavior must change? How will we know the operating result is better? If those answers are unclear, the technology is ahead of the operating model.

The practical question is not “Where can we use AI?” It is “Which operating decision should improve, who owns it, and what evidence will show that it did?”

What to do next

Choose one material business process. Name its owner, the decision or handoff that needs to improve, the controls that matter, and the measure that will prove movement. Then decide whether AI is the right intervention.

Bring the initiative that needs an operating model.