​

Are CEOs innovating around artificial intelligence (AI), or are they panicking?

On the surface, it looks like executives are doing what they’ve been repeatedly told to do when it comes to emerging technologies—pilot, experiment, drop the fear, embrace the future. According to a new IBM survey of CEOs, the majority of top leaders are all-in on AI: two-thirds of them say they are “comfortable using AI to help inform major strategic decisions,” three-fourths now have a chief AI officer, and nearly half—48%—say they believe AI will be able to handle some decisions without human intervention by 2030.

But how much of that activity is born of genuine enthusiasm for AI is an open question. According to a Boston Consulting Group (BCG) survey, 61% of CEOs say their boards are rushing into AI. Almost 40% say their boards don’t have an “informed view” of the technology, and approximately one-third of respondents say their boards “overestimate the human capabilities that AI can replace.”

The pressure to do something around AI isn’t just coming from boards. According to a report from the technology firm Dataiku, 80% of CEOs say they feel their jobs are on the line if they can’t deliver clear wins around AI this year. Echoing the BCG report, around 60% of CEOs say they’re being nudged by their boards to do something around AI. As the report puts it, “AI is now a standing agenda item, not a strategic sidebar.”

Should it be? The question will be different for every organization. But taken together, the three reports suggest that a lot of leaders are dedicating a lot of time to AI, while holding a fair amount of skepticism about its value. The seductive qualities of the technology are easy to identify—low barrier to entry, opportunities for robust automation, and smarter and quicker analytics. All of which should add up to more revenue and/or cost savings, with staffers’ brains now free to focus on more interesting and complex tasks.

But those revenue numbers aren’t there yet, and the Dataiku report argues that executives are less enthused about AI when they’re outside the boardroom. Indeed, they’re actively worried: 57% say that an AI failure “could trigger a crisis that erodes customer trust or damages their brand.” Small wonder they’re worried: The report notes that a lack of data governance is risking “data exposure, inconsistent outputs, and unmanaged dependencies.”

That’s not an argument for CEOs to pull the plug. But it’s a prompt for them to be honest with their stakeholders—and themselves—about where they stand in terms of AI governance, risk tolerance, and strategy. A board member saying, “we should do something with AI” in 2026 is no smarter than one saying “we should do something with the internet” in 1998. Yes, you should, probably. But what that will be is a function of understanding what AI can provide and whether it’s in an organization’s capacity to manage—and it may be that you’ll ultimately move slower and less expensively than the hype.

This article originally appeared on AssociationsNow.com. Reprinted with permission. Copyright ASAE: The Center for Association Leadership (May 2026), Washington, DC.

Mark Athitakis, a contributing editor for Associations Now, has written on nonprofits, the arts, and leadership for a variety of publications. He is a co-author of “The Dumbest Moments in Business History” and hopes you never qualify for the sequel.