AI is the CEO's job.
AI has long since arrived in the company — it just often isn't led yet. Whoever doesn't lead it breeds shadow AI and still carries the legal responsibility — NIS2, EU AI Act, GDPR.
In the conversations I'm having with CEOs right now, AI is no longer a future topic. It's here. People are using tools, individual units are experimenting, somewhere a pilot is running. What's missing isn't the technology. What's missing is the decision about what all of it is actually meant to serve — and who makes it.
Instead, AI gets delegated to IT or pinned to a single tool. The result: pilots without direction, the feeling of running behind, busywork instead of impact. Not because no one is doing anything. But because no one is leading.
The question isn't whether AI is in the company. It's who leads it.
Why AI stalls in IT.
IT can provision tools, secure them, integrate them. What it can't do: decide which use cases carry the biggest leverage for the business, which risks the leadership wants to carry and which it doesn't, who in the company is responsible for what. These aren't technical questions. They're leadership questions.
As long as they stay unanswered, it isn't impact that grows but the sprawl: shadow AI, unclear data flows, tools without a line. And the liability stays at the top regardless. EU AI Act, data protection, NIS2 — they address the leadership, not the tool.
AI governance isn't an IT question. It's a leadership matter.
The more honest sequence.
Whoever delegates AI to IT delegates a decision they can't delegate — the responsibility stays. The more honest sequence is the reverse: first the leadership decision about what AI serves, then the tool choice. Not the other way around.
Facilitating AI as a leadership question.
Three consequences.
One shift, three levels. What changes in concrete terms — for you, in the team, in the organisation.
The question isn't “Which tool do we buy?” but “Where does AI carry real leverage in our business — and where is it only hype?” The consequence for the coming week: name three use cases where AI genuinely makes a decision easier, and make one of them the CEO's job. KI-Kompass is a method, not a tool: it clarifies the direction before you pick the tools.
“Who takes care of AI?” is the wrong question. The right one is: “Which rules, roles and risks do we carry together?” The consequence: a clear read on your own AI maturity, prioritised use cases and first guardrails — agreed together, not pushed off to IT. That's how AI becomes a leadership matter instead of a one-off initiative.
Instead of “How many tools do we have in use?”, the more honest question: “Which of them actually contribute to the business?” The consequence: a 90-day plan with named owners instead of scattered pilots. At millionsteps we build exactly this bridge from the decision into practice — so that led AI becomes a load-bearing system, not the next tool in the sprawl.
The measure isn't more AI.
Growth that doesn't carry is just speed.
More isn't growth. That holds here too: more tools, more pilots, more speed aren't the same as progress. What counts is whether what you put into use carries in the business.
AI amplifies both — an organisation that's led, and one that's merely busy. The difference doesn't come from the technology. It comes from the leadership.
Where does AI carry
real leverage for you?
If this description comes closer to your situation than you'd like, let's work it through in a 30-minute AI orientation: where AI really counts in your company, which risks you have to lead — and what a sensible first step would be. As a leadership decision, not a tool question.
Book a discovery call →Check for yourself: AI maturity check ·
Shadow AI · 5-point check
The format for it: KI-Kompass ·
The focus area: AI in the responsibility of
leadership