The AI Adoption Radar: Why I Built a Tool That Refuses to Give You a Score

By Martin Bergmann | AI Project Lab | August 2026

PMI’s new AI Standard makes something explicit that most organizations have been quietly avoiding: the project manager is the accountability checkpoint for AI on their projects. Not IT. Not governance in the abstract. You.

That raises a question most self-assessments don’t actually answer: are you using AI well, or are you just using AI?

The Question Everyone Gets Wrong 

Almost every “AI maturity” check asks the same thing: how often do you use it? That’s the wrong question, or at least a dangerously incomplete one. Frequency tells you whether AI has become a habit. It tells you nothing about whether that habit is any good.

I see this constantly in PMO conversations. Someone is fully embedded — AI in their status reports, risk logs, and stakeholder emails—and genuinely proud of it. But ask a follow-up: is there a real human-in-the-loop check before that output goes out? Could you defend that deliverable if a stakeholder pushed back on it? The answer is often a pause that says more than the frequency number ever could.

The reverse happens too. Careful, deliberate people who review everything twice — and have barely integrated AI into their actual workflow. Safety without adoption isn’t a virtue. It’s falling behind carefully.

Two Things Have to Be True at Once

The PMI AI Standard’s Human-in-the-Loop principle doesn’t let you off the hook either way. It requires both: AI genuinely embedded in how you deliver, and real oversight, quality control, and disclosure around how you use it. Integration and accountability aren’t in tension. They’re both required, simultaneously, and most people are strong on one and thin on the other. That’s the gap I wanted a tool to actually show — not just tell people about.

Why There’s No Overall Score

Here’s the design decision I keep coming back to: the AI Adoption Radar deliberately does not give you a single composite number. It would have been easy to add one — average the eight dimensions, slap a label on it, call it a day. I built it that way at first, honestly. Then I realized it was undermining the entire point. Two people can land on the exact same average and be in completely different positions: one strong on integration and dangerously thin on accountability, the other careful but barely using AI at all. A single score erases exactly the distinction that matters. It reduces the diagnosis to a number that feels precise but tells you nothing about where to actually focus.

So the tool maps eight dimensions across two groups — Integration (Frequency, Depth, Tool Spread, Team Diffusion) and Accountability (Oversight, Confidence, Quality, Transparency) — and shows you the shape. Not a grade. A picture of where you actually stand, and where the gap is.

Try It Yourself

It takes about 90 seconds. Rate yourself honestly across the eight dimensions, watch the shape form, and see which side — integration or accountability — is thinner than you’d like.

Take the AI Adoption Radar →

If you want to act on what you find, there’s a companion playbook: concrete, doable-this-month actions for each dimension, plus a worksheet to turn your weakest one into an actual 30-day plan instead of a good intention.

---

Martin Bergmann is the Director of an IT PMO and founder of the AI Project Lab. He writes about AI, project management, and the future of intelligent work.

One idea per week, no hype, always something you can use.