Boards are clear. The C-suite is running AI.
In a new Pearl Meyer survey of 108 executives and board members released on Wednesday, 90% of board members said responsibility for leading artificial intelligence effectively belongs with the C-suite and their direct reports—essentially all the most-senior executives within a company.
Inside the C-suite itself? Executives are pointing in four different directions.
Corporate leaders surveyed in February and March by Pearl Meyer, an executive compensation and leadership advisory firm, splintered into different camps on the question of who among them actually owns AI. The results showed 32% said the C-suite as a group is accountable for AI strategy; 22% pointed to the group one level below the C-suite; 27% pointed to individual business leaders; and 17% said AI sits with functional heads like HR, finance, and legal.
As companies move from piloting AI toward enterprise-level rollouts, this divergence of views raises an important question with real-world consequences: If something goes wrong, who is responsible for catching it before it goes public?
Pearl Meyer’s data also points to a broader problem underneath the AI governance gap. Boards and executives don’t agree on how cohesive their leadership teams actually are, on whether strategic priorities are traveling down the organization, or even on which factors matter most to scaling AI in the first place. For most companies, those gaps existed before AI exploded as a usable workplace tool. What’s changed now is that AI is the most visible live wire running through them—and the most likely to result in a public gaffe if it’s managed badly.
According to Brad Jayne, a principal at Pearl Meyer and one of the survey report’s authors, AI itself isn’t creating new problems. The ownership split is the symptom of a problem that’s been hidden inside C-suites for years.
“Your leaders don’t know how to be a team,” said Jayne. “C-suite teams, they can all perform on their own, but collaborating and actually figuring out how to work effectively as a team isn’t there. So when you see these big changes externally that they have to react to—I think AI just shines a light on something that was already there.”
The rest of the survey showed similar disconnects between boards and different groups of executives.
Pearl Meyer found 100% of the directors in the survey believe their senior team is a cohesive enterprise unit. Only 66% of C-suite executives agreed, while 34% said they didn’t believe their team worked well together.
A similar pattern held on how decisions cascade down. On communication priorities, 100% of board members said decisions made by the senior leadership team translated into clear priorities, versus 78% of C-suite respondents. When Pearl Meyer narrowed the question further and asked whether leaders two levels below the C-suite can clearly and consistently explain the company’s top strategic priorities, only 54% of C-suite executives said yes.
That means that even among executives who think the strategy is clear at their own level, nearly half aren’t confident it has accurately traveled down to executives who do the nitty-gritty work that would be part of an AI rollout.
Jayne said boards tend to get sold the top-line AI story and don’t press far enough on the operating reality beneath it.
“The board can swoop in, hear the story, invest in AI, support that, but then maybe they don’t spend enough time fully understanding where that might impact the organization,” he said. “The storyline hits the top level, but then they’re just not really sure how they’re going to go about it.”
In other words, the C-suite might be telling the board, “We’ve got this,” said Jayne. “Then internally the C-suite says, ‘We have no idea how we’re going to do this.’”
‘Just start using it’
Looking at the way AI has been piloted and positioned inside companies shows that it has been somewhat laissez-faire. Once the basic guardrails are in place, Jayne said, the message from senior leaders on AI tends to be a single command: Go.
“The message from leadership is often, ‘Just start using it,’” he said. “And they miss the rest of the story, which is, ‘We’re not exactly sure where to use it.’” Whether employees are using AI and how effective they are is also unclear, as is whether they are actually more efficient, he added.
The data supports him. When Pearl Meyer asked respondents to name the most important factors impacting their company’s AI preparedness, boards and executives chose almost entirely different responses.
Board members focused on ownership, with 45% saying clear executive ownership and decision rights was a top-three factor in being ready to deploy AI. Only 22% of C-suite respondents agreed. Other executives zeroed in on the workings underneath with 49% pointing to data quality, infrastructure, and security as a top factor, compared with only 18% of board members.
Peter Thies, a managing director at Pearl Meyer and coauthor on the survey report, said each side’s answer reveals how they see the business.
“The C-suite’s not that concerned about who owns [AI], because a lot of people actually have something to do with it,” said Thies. Inside companies, AI might touch almost every function including tech, HR, finance, legal, and individual business units. Distributed responsibility seems less like a governance gap than a description of how AI is working. But for board members who see the organization from the outside looking in and hear about it straight from the CEO and top executives, that reads as nobody is in charge.
When it comes to data quality, on the other hand, the split runs in the opposite direction, the survey showed.
“C-suite, they’re all over that one,” said Thies. “And yet the board doesn’t see how important [data quality] is to the company.”
Industries that will struggle the most with discrepancies at the top around who owns AI and how it is being deployed operationally are likely in sectors where leadership tenure is the longest and where culture changes are hardest, said Jayne.
“Finance or banks, maybe insurance companies, places where people can often have a very long tenure—it’s difficult to move the needle,” said Jayne. Financial services was the largest industry represented in the Pearl Meyer sample, at 34% of respondents.
Some 71% of executives told Pearl Meyer that success over the next 12 to 18 months will depend on fixing internal processes and cross-functional coordination—not on AI itself.
“Leadership systems are not evolving fast enough to support either strategy or AI,” the report concludes.
None of this is happening in a vacuum. Companies including Block, Meta, and Oracle have announced AI efficiency gains as the reason for workforce cuts, and the stock market rewarded them.
That reaction creates pressure on every other CEO to deliver the same story, whether the AI is actually doing the work or not.
“At times I see AI being used as the reason for things that may have come about anyway,” said Jayne. And the more pressure there is to use AI as a justification for efficiency gains, the more pressure builds to show real results in terms of key performance indicators that make sense to internal employees and external shareholders, he said.
Pearl Meyer’s data shows 40% of companies are still piloting AI, and 31% are experimenting or using it on an ad hoc basis, not because it isn’t useful, but likely because the leadership teams needed to deploy it at scale aren’t in agreement on how to do it and what matters most.
“Maybe wheels are spinning a little bit,” Jayne said. “Are we about to shoot off down the road? I don’t know. But it’s a little slower to get going than I thought it would be.”
Pearl Meyer surveyed 108 respondents from 40 public companies, 58 private companies, and 12 nonprofits/government entities.
