Chapter 4: AI Readiness

What the instrument measures

The AI Readiness Assessment scores twelve statements across four dimensions. It deliberately does not measure tool adoption, spend or pilot count, because those measure activity rather than readiness.

• Placement. Whether the leader can name where AI would change business outcomes, and whether AI work starts from an outcome rather than from a vendor demo.
• Restraint. Whether the organisation has decided what it will not automate yet, checks whether process and data are ready before scaling, and is willing to say not yet.
• Ownership. Whether AI is owned at leadership level rather than handed to IT or a vendor, whether people still own the important calls, and whether a failed initiative would have a named human owner.
• Momentum. Whether AI is treated as an ongoing capability rather than a project, whether loops are short and honestly reviewed, and whether a concrete result could be shown to a board.

Each dimension is defined in full in the leadership metric library.

Why the sample is small

Three reasons, stated plainly because they affect how quickly this chapter becomes reportable.

1. The instrument is the newest of the four. Its fields were created on 27 June 2026, so this is the first full quarter in which any completion was possible.
2. It is the longest of the four assessments and sits on a separate page from the main assessment hub, so fewer visitors reach it.
3. Through most of the quarter the assessment pages carried no analytics container, which was fixed on 13 September 2026. That did not affect data capture, but it did mean the funnel into this assessment went unmeasured and therefore unimproved.

All three are fixable and two are already fixed. On the current run rate the floor is reached during Q4.

What will be reported when the floor is cleared

So that the next edition cannot be accused of choosing its cuts after seeing the data, here is what will be published when n reaches 30, and nothing else:

• Distribution of readiness level across respondents.
• Mean and median score for each of the four dimensions, on the same 0 to 100 scale.
• The lowest scoring dimension, and the share of respondents for whom each dimension was weakest.
• The share of respondents below the readiness threshold.

No cut with fewer than 30 responses will be published, including within these four.

What the other three chapters already suggest

This is interpretation, not data. The three reportable chapters describe a population that is strong on relating to people and weak on setting direction and persuading. The Ownership dimension of AI readiness is a people and accountability question, which this population would be expected to handle well. Placement and Momentum are direction setting questions, which is exactly where the other three instruments show the gap. If that pattern holds, AI readiness in owner led companies will turn out to be limited by direction rather than by technology or by trust. The Q4 edition will test that, and will say so if the data disagrees.