From AI-using to AI-native.

Stop one

Start with an honest number.

The AI-Native Readiness Checklist is seven dimensions, three statements each, scored out of 21. It takes about five minutes and it is the same set of questions I work through with executive teams. Score it the way things are today, not the way the roadmap says they will be. A flattering number helps no one (I cannot see your score, so if you embellish, you are lying only to yourself).

Stop two

Read your score the way I do.

The number is the thermometer, not the diagnosis. Here is how I read the three bands.

Notice what that means. A 7 and a 19 are not the same distance along one road. They are different problems that need different next moves. The score tells you which conversation you are actually in. Then go back and find your lowest-scoring dimension. That is almost always where the next move lives, not the place you are already proud of.

Stop three

The trap: activity is not value.

Most AI programs measure the wrong thing, and they measure it enthusiastically. Number of pilots. Seats licensed. Prompts run. Hours saved in a survey nobody audits. These are activity metrics. They go up and to the right, they make a beautiful slide, and they tell you almost nothing about whether the business changed.

Here is the tell. Can a CFO or COO point to a business metric, cycle time, throughput, quality, margin, risk, a customer outcome, that AI has measurably moved? Not "saved us time." Moved. In the numbers that already run the company.

If the answer is a pause, you do not have a technology problem. You have an activity-versus-value problem, and it is the most common one I see. The work got faster in places. The company did not change shape. Those are not the same thing, and only one of them shows up in the P&L.

Stop four

You cannot bolt your way to native.

Here is what the checklist is quietly measuring underneath all seven dimensions. The org chart, the approval gates, the metrics, the handoffs were all built for a time when intelligence was scarce and human. Add AI to that and you get a faster version of the old company. Briefly impressive. Structurally unchanged.

Bolting AI onto your old operating model is like putting a jet engine on a horse cart. You do go faster, right up until the cart explains (at speed) why it was never built for this.

Going from AI-using to AI-native is not a tooling change. It is rebuilding how four things actually flow:

• Work — redesigned from the customer outcome backward, not sped up in place.
• People — roles, judgment, and handoffs redrawn for humans and AI working together.
• Governance — lanes that let good work move, not a single gate that blocks it.
• Value — measured in business outcomes, so the change is defensible to finance.

Those four are the engine under the seven dimensions you scored. When a team is stuck, it is almost always because one of these four is still running on the old design while the others tried to move.

Stop five

The conversation, if you want it.

If the checklist surfaced something real, you have two honest options.

The first is to keep reading. Trouble Worth Making works these ideas in public, roughly weekly, and that is genuinely enough. Most people should just do this.

The second, if the gaps feel urgent and expensive, is to talk it through directly. I run focused AI-native readiness conversations with leadership teams. We take your actual situation, not a generic maturity model, and find the one or two moves that would change the most. No deck, no pressure, and no offense taken if the answer is "not yet."