The AI leadership assessment.

An AI leadership assessment measures how a leader leads when machines can think, which is a different thing from how fluently they use this quarter’s tools. The gap most organizations have is personal before it is technological, and the usage dashboard can’t see it. You bought the licenses, and usage tells a different story, because adoption and compliance look the same from a dashboard.

This page covers what the AIAdoptionindex™ measures, what the read-out shows, what it is honest about not being, and how to use the baseline to pick a pilot. There is a two-minute self-check below if you want to start with the week you actually live.

What the assessment measures

The AIAdoptionindex™ scores behavior frequency across The 5 Disciplines of AI Adoption™. Define: naming the outcome and the workflow you are actually changing, the actual week rather than the job description. Discover: finding where the work really happens, and asking what you can do given what AI can now do. Design: redesigning the workflow around the capability, not the demo. Develop: preparing the people who have to execute the pilot, and instrumenting it. Demonstrate: proving value in the language of the business before anyone asks.

It is not a knowledge test. It measures how often the behaviors behind each discipline actually happen, which is harder to flatter than a conversation about AI strategy.

What it is honest about not being

The AIAdoptionindex™ is a self-assessment. It is not validated, it is not multi-rater, and it does not claim to be. Its job is a baseline per discipline at the start of the program and a re-measure at the end, so you can measure the shift and see where leaders moved, not just whether they attended. That is a modest claim, and it is the right one for a field where the tools change every quarter and the operating discipline is the only thing that holds still.

What the read-out usually shows

Consider a head of customer support at a SaaS company with a copilot on every seat and a dashboard that says adoption is fine. Her index scores high on Define and Design and near zero on Develop and Demonstrate. She has a well-designed pilot for ticket triage that the agents quietly route around, because nobody asked whether they could be bad at it without being punished, and no evidence chain exists to tell her whether it works. The read-out does not say her design is wrong. It says implementation failure is usually a readiness failure, not a design failure, and names the discipline where readiness lives. Her next move is a stakeholder support plan with the most hesitant agents on it, and a 90-day proof plan with a leading indicator the agents chose.

What to do with the result

Work the lowest discipline first, on one bounded workflow. AI leadership training has each leader run the five disciplines on a real pilot, with a deliverable at each step: the partnership audit, the possibility map, the pilot plan, the stakeholder support plan, and the 90-day proof plan. Then re-take the index. The movement per discipline is the measure. The pilot’s evidence chain, leading indicators for confidence, lagging indicators at sixty to ninety days, and the story that gets you the culture rather than the budget, is the result the business recognizes.

After Demonstrate, you return to Define with data. Courses end. Disciplines compound.

Where to start

Start with the self-check above against your current quarter. Then take the AIAdoptionindex™ and bring the one workflow you would redesign if you had to pick today. The lowest discipline is where the program starts, and the re-measure at the end tells you whether you moved.