The Development
Board Intelligence, EMEA’s largest board technology and advisory firm, released the Summer 2026 edition of its Board Value Index on June 11, drawing on responses from 405 non-executive directors, CEOs, and CFOs at organizations with revenues above 50 million pounds across the United Kingdom, United States, Nordics, and Middle East. The headline finding is blunt: 86 percent of directors say overly rigid or inconsistent decision-making frameworks contributed to a delayed, rushed, or poor decision in the past six months.
The study lands at the precise moment boards are being asked to govern artificial intelligence. More than four in five directors, 84 percent, report that their boards are now debating which decisions should remain human-led and which can be delegated to AI. Yet 40 percent believe their own board will require little or only incremental change over the next five years. The gap between the disruption boards see in their organizations and the disruption they expect in their own conduct is the real story.
Why It Matters to the Board
This is not a technology problem. It is a decision-quality problem surfacing at the worst possible time. Boards are being measured less on oversight and compliance and more on the speed, clarity, and quality of their judgment under uncertainty, and the Board Intelligence data shows the operating system most boards run on was built for a slower era.
The obstacles directors cite are structural and therefore fixable: 34 percent point to decision-making frameworks themselves, 32 percent to unclear roles between the board, executives, and committees, and 29 percent to the quality of information reaching the board. When a board cannot say cleanly who decides what, layering AI on top does not create leverage. It compounds confusion.
For a Fortune 500 director, the implication is direct. The same ambiguity that produced a rushed decision last quarter will govern how the board approves AI deployment, model risk, and the boundary between human and machine authority this quarter.
The Risk If You Wait
The danger is not that boards reject AI. It is that they adopt it inside a decision architecture they already concede is flawed. Pippa Begg, Board Intelligence’s CEO and co-founder, framed the core question as where human judgment should end and AI should begin. A board that has not answered that question in writing will answer it by default, decision by decision, with no record of why.
Delay also carries a documentation cost. Forty-one percent of directors say their boards spend at least half of meeting time looking backward at past performance rather than forward at strategy. A backward-looking board confronting a forward-moving technology will consistently ratify management’s AI choices after the fact rather than shaping them in advance, which is the definition of oversight failure in a future enforcement review.
What Other Boards Are Doing
The leading boards in the data are separating themselves on two moves. First, they are putting the human-versus-AI decision boundary on paper, treating it as a governance artifact rather than a hallway conversation. Among the surveyed population, the boards reporting strong value creation are the ones redesigning how they process information and challenge assumptions, not simply buying tools.
Second, they are confronting the innovation gap inside their own ranks. The survey found 27 percent of CEOs and CFOs say their board strongly enables innovation, but only 12 percent of non-executive directors agree. High-performing boards are closing that perception gap deliberately, equipping independent directors with the visibility and fluency to challenge management on AI rather than defer to it.
The Governance Question
Every director should bring one question to the next meeting: can we produce, today, a written statement of which categories of decision in this enterprise may be made or materially influenced by AI, who owns each one, and how the board reviews them? If the answer is no, the board is exposed regardless of how sophisticated its technology stack appears.
The follow-on question is sharper. If 86 percent of our peers admit their frameworks produced a poor decision this year, what specific evidence do we have that ours did not?
Intelligence Bottom Line
The Board Value Index confirms a divergence that will define the next governance cycle: organizations are being remade by AI while the boards overseeing them expect to operate largely unchanged. That posture is no longer defensible. The boards that hold up under scrutiny will be those that treat their own decision-making process as the first system requiring AI-era redesign, not the last.
The mandate is concrete. Document the human-AI decision boundary, clarify role ownership, and shift meeting time from retrospective review to forward governance before the next major AI decision forces the issue. The data suggests most boards have roughly one cycle to act before the gap hardens into a liability.