Picture a CEO at a company like yours. Her CIO is one of the best hires she ever made. He knows every system, every vendor, and every contract, and he keeps the lights on through weather that would sink a less careful shop. At the board meeting, a director asks what AI is doing for the company. The answer is a list of pilots, a license count, and a plan for next year. Nobody in the room is wrong. Nobody in the room is satisfied, either.
If that picture is uncomfortable, it is probably because it describes a seat and not a person.
If that picture is uncomfortable, it is probably because it describes a seat and not a person. The CIO’s seat now asks for two jobs at once: run the platform the company depends on today, and move AI forward at the pace the CEO and the board expect. This article looks at what Gartner’s newest numbers say about how often one seat delivers both, and what a CEO can do when the honest answer is “not yet”. None of it requires indicting the person currently holding the role.
The Number
At its October 2026 IT Symposium, Gartner analyst Alicia Mullery said that only 11 percent of CIOs are what Gartner calls AI Vanguards, meaning they deliver strong business outcomes from AI and their CEOs perceive high value in it. A companion release the same week reported that only 13 percent of CIOs said their enterprise had realized significant or transformational value from AI so far. The 13 percent comes from Gartner’s 2027 CIO and Technology Executive Agenda survey of 2,002 CIOs and technology executives, fielded from May to July 2026. We read the 11 percent as coming from the same research, although the session summary does not say so.
Four cautions belong next to these figures. First, “vanguard” is Gartner’s label, and it blends outcomes with the CEO’s perception of value, which is not the same as audited results. Second, the respondents are CIOs and technology executives describing their own enterprises. Third, what Gartner has published does not break the results out by company size, so a mid-market CEO should not assume the number is better or worse for a company like theirs. Fourth, and most important, the traits that follow are correlations. They tell you what vanguards are more likely to do, not that doing it makes a company a vanguard.
What Vanguards Do Differently
The pattern is not about tools. Gartner found that organizations with guidelines for deciding when an AI pilot should not move to production were 2.3 times more likely to be vanguards, and those with guidelines for abandoning low-value AI already in production were 2.6 times more likely. Only 31 percent and 17 percent of CIOs, respectively, said they had those practices. Organizations with explicit plans to redesign roles for a workforce of people and AI agents were 2.4 times more likely to be vanguards, and only 4 percent of CIOs had such plans.
Look at what those practices have in common. Stopping a pilot, retiring a tool that is not paying, and redesigning how work gets done are not technical tasks. They are decisions about how the company operates, and they require business leaders who are willing to walk away from an investment, which Mullery said directly. Gartner also says CIOs need to help their organizations govern AI economics across the full lifecycle, not simply approve the initial purchase. That is a different job from keeping the platform healthy, and a larger one.
Remember the limit: these are likelihoods. A company can have every one of these practices and still disappoint, and a company with none of them can have a good year.
Why the Seat Is Squeezed
Gartner’s 2027 planning guidance, published August 31, 2026, projects average IT budget growth of just 3.7 percent in 2027, while funding for agentic AI is expected to rise an average of 31.8 percent. Those are different bases, so do not divide one by the other, but the direction is clear: the AI line is growing far faster than the budget that has to pay for it. In the same research, 58 percent of CIOs reported growing pressure to deliver cost improvement from AI, 51 percent expected AI to increase the total cost of ownership of technology, and 44 percent cited insufficient technical talent. Gartner’s own description of the situation is that many CIOs are operating within budget, governance, and workforce structures that were not designed for the pace and scale of AI.
The platform has to run. It has to run first.
Now the part that is our view and not a finding. We have not found a published split of CIO time between running technology and changing it that we would trust, so we will not quote one. Our observation is simpler. When a technology leader’s days are filled with renewals, outages, security reviews, and audits, the AI agenda gets whatever is left over, and what is left over is rarely enough. The platform has to run. It has to run first.
The CEO side of the table is moving too. In Gartner’s 2027 CEO survey of 203 CEOs and senior business executives, fielded in April and May 2026, 80 percent said technology has become strategy, 44 percent said they struggle to demonstrate financial returns from AI, and 63 percent named radically redesigning the operating model as a top move to accelerate AI returns. It is a small global sample, so treat it as direction. A Gartner analyst warned that CIOs who remain downstream implementers risk being left out of the decisions that define competitive advantage. For a CEO, the seat has stopped being a support function.
A Problem With the Seat, Not the Person
A CIO can be excellent at the job as it was defined five years ago, and the job has changed. Gartner notes that AI causes roles to compress and expand at once: broader roles, fewer role types, more demand for technology talent. The CIO’s seat is the clearest example. It now includes choosing which AI bets to make and which to stop, redesigning work alongside business leaders, governing cost across the lifecycle, and keeping everything else running.
That is why a CEO can say this out loud without indicting anyone: this seat now asks for two jobs, so let’s make sure it has what it needs. Nothing in that sentence is a verdict on a person. And the incumbent’s knowledge of your systems, vendors, and history is an asset the AI agenda depends on, not an obstacle to it. The question is whether the seat has the capacity and the experience to do the larger job, not whether the person in it deserves to have it.
Five Questions a CEO Can Ask This Week
These are our questions, not Gartner’s, and they are meant for a conversation and not an audit. First, can your CIO name the AI initiatives that will be stopped if they miss their numbers, and the date on which that decision gets made? Second, does someone own AI results in business terms such as revenue, cost, or cycle time, and is that written down somewhere? Third, ask your CIO what share of the last quarter went to keeping things running and what share went to moving AI forward, and listen to how easy the answer is to give. Fourth, do you have written criteria for when a pilot does not go to production? Fifth, has anyone beside your CIO, an adviser or a peer, led a company of your size through a change like this before and told you what comes next?
If three or more of those answers are shaky, the seat is probably short on capacity, experience, or both. Our article on proving AI works when nobody will work without it covers the measurement side of the first two questions.
Four Moves, Lightest to Heaviest
The first move is coaching: give your CIO a peer who has done this before, a former CIO who can speak plainly about the sequence. It is cheap, it respects the person, and it works when the gap is experience and not time.
The second is clearing the runway: take operational load off the CIO’s desk through a strong deputy or a managed provider, so the AI agenda gets protected hours. Be careful with this one. Concentrating more of your operations in a single provider has its own tradeoffs, which we covered in Want to Direct the AI Dividend? You Must See the Tradeoffs Clearly.
The third is adding an innovator beside the CIO: a seasoned interim or fractional technology leader who owns the AI agenda for six to twelve months, reports to you, and works with the CIO and not around them. The fourth is reshaping the seat itself, putting a seasoned innovator in the CIO role and giving the operations leader the platform below it. Which move fits depends on facts about your company that no survey can supply.
Our Argument, Stated Plainly
This part is argument and not evidence, and we want you to read it as one. We believe that if your CIO is not a vanguard, or is too occupied running the platform to move AI forward at the pace you now need, then getting a seasoned innovator into that role, or alongside it, is likely to do more for your company’s future than any other role you fill this year.
The seat touches every function, so improvement there multiplies and weakness there does too.
There are three reasons. The seat touches every function, so improvement there multiplies and weakness there does too. The vanguard differences are decisions about stopping, redesigning, and governing, and a person who has made those decisions before in other companies carries judgment that cannot be hired in as a tool. And the clock is running: the same Gartner keynote release found that 71 percent of organizations expect to deploy agentic AI within twelve months, while only 21 percent have the governance capabilities in place, a gap we covered in Less than a Third of AI Agents are Governed.
Two clarifications. This is not a split between IT and AI. The platform and the AI agenda are one company, built on the same data, security, and integrations, and a seat that treats them separately creates the gaps governance exists to close. A seasoned innovator in the sense we mean has run platforms too. And Gartner has not tested whether adding or changing a leader raises a company’s odds, because it measures practices and not people. The case rests on our judgment, and you should weigh it that way.
What This Will Not Do
We think a seasoned leader improves your odds. It does not guarantee you become a vanguard, a label that is Gartner’s and that depends partly on how your CEO perceives value. It will not work if the company will not stop projects, since stopping criteria need business partners willing to use them. And the evidence has the limits described above: correlation, self-reported results from CIOs and technology executives, no size breakdown, and a small CEO sample. If your CIO already answers the five questions well, you may be closer to the 11 percent than you think, and the best move may be to protect their time.
Where to Start
Begin with the five questions, and answer them honestly before you decide anything about anyone. If you would like an outside read, an independent assessment gives you the state of your technology and AI in your own terms, from people with no stake in whose seat is whose. Innovation Vista keeps a bench of more than 450 former CIOs, CTOs, and CISOs, and the right match might be a coach, an interim CIO, or a virtual CIO working beside the leader you already have. Treating the CIO’s seat as the highest-leverage one in the company, as we argue it is, does not indict the person in it. It asks only that the seat be built for the job it now has.
