The CEO Owns the AI Strategy Alone. The Board Just Noticed.
Seventy-two percent of CEOs now say they are the main decision maker on AI strategy at their company, according to BCG’s 2026 AI Radar survey of more than 1,250 executives. That share has roughly doubled in a single year. Company AI spending is set to double too, climbing toward 1.7 percent of revenue.
Read that pairing slowly. The money is moving fastest exactly where the judgment is most concentrated. One person is deciding what your company builds, buys, and bets on with AI, and that person is busier than ever running everything else the company does.
Why Is AI Strategy Landing on One Desk?
Speed explains most of it. Nobody wants to be the company still forming a committee while a competitor ships. A CEO who waits for consensus looks slow. A CEO who just decides looks decisive, at least until the decision goes wrong.
Ownership explains the rest. AI does not sit cleanly inside one function. It touches product, hiring, finance, customer service, and the factory floor, all at once. When a decision belongs to everyone, in practice it belongs to no one, so it drifts upward until it lands on the one desk where “whose job is this” stops being a question. Half of CEOs in the same research now say their own job depends on getting AI right. That is not a group they are eager to hand a piece of the decision to.
The Chief AI Officer Title Did Not Fix This
Companies noticed the gap and tried to patch it. The share of companies with a Chief AI Officer jumped from roughly a quarter to more than three-quarters in a single year. It looks, on an org chart, like the concentration problem got solved.
It did not. A title is not authority, and a new box on a chart is not a shared judgment. Most Chief AI Officers were hired into a coordination role, not a decision-making one. The CEO still makes the call. The new hire mostly makes the case, or cleans up after it. Adding a title without moving the actual decision rights is the org-chart version of hanging a picture over a crack in the wall.
What Boards Are Asking Now
The tone from boardrooms has shifted from encouragement to a quieter kind of concern. Analysts covering this year’s CEO surveys describe boards pushing their chief executives to slow down and re-anchor, not because AI is a bad bet, but because one person’s judgment is now standing in for an entire leadership team’s.
That is a governance question dressed up as a technology question. Every board already knows what happens when one executive’s read of a market, a merger, or a risk goes unchecked. This is the same exposure, just newer and less familiar, so it took longer to name.
The Single Point of Failure Problem
Here is the part that should worry you more than a bad quarter of AI spending. A single decision maker has a single set of blind spots. Whatever the CEO does not see, misjudges, or simply has not had time to learn, the entire company inherits. There is no second reader. No colleague in the room who has spent real time in the part of the business where the AI tool will actually land.
This is the same failure mode this site keeps circling back to, in a district’s business office and in a Fortune 500 boardroom alike: an invisible risk sitting in plain sight because only one person was ever asked to look at it. Making that risk visible does not require a bigger AI budget. It requires more eyes on the same decision before it ships.
A school superintendent who single-handedly picks the AI tool for every classroom carries the identical risk as a CEO doing the same thing for a workforce of thousands. The scale differs. The blind spot does not.
Neither leader is doing anything wrong on purpose. They are doing what every fast-moving organization rewards: deciding quickly, owning the outcome, and moving on before the meeting runs long. The problem only shows up later, when the decision meets a part of the business the decision maker never actually worked in.
What Actually Distributes the Judgment
None of this argues for slowing AI adoption down. It argues for widening who is in the room before a decision gets made, not after it needs defending.
Name the categories, then name the owner of each one. Not every AI decision belongs to the CEO. A decision about customer-facing AI belongs with whoever owns the customer relationship. A decision about AI in hiring belongs with whoever will answer for a biased outcome. Write the list down. Most companies have never done this on purpose.
Bring the leadership team into the same room, not a memo. A CEO who announces an AI direction in an all-hands has told the company what happened. A CEO who works through the tradeoffs out loud with the people who run finance, HR, and operations has actually shared the judgment. Those are not the same event, and only one of them survives the first hard question.
Ask what the CEO cannot see. This is the uncomfortable one, and the one most rollouts skip. Every leader has a function they understand least. Naming it out loud, in front of the team, is the fastest way to find out where the company’s biggest AI blind spot is actually hiding.
Measure whether the decision right actually moved. A Chief AI Officer on the org chart proves nothing on its own. Ask, six months later, who made the last three real AI calls. If the answer is still one name, the title was decoration.
Seventy-two percent of CEOs now own the AI strategy alone, and their boards have started to notice what that concentration actually costs. The fix is not a new hire or a new dashboard. It is a leadership team that has actually sat in the same room, argued through the same decisions, and left knowing who owns what.
That is the work The Spotlight Machine’s leadership training does with executive teams directly: not a briefing to the CEO alone, but a room where the whole leadership team builds the shared judgment an AI strategy actually needs.