Nearly 90% of CEOs Think Employees Trust Them. Only 46% Do.

Nearly 90% of CEOs believe their employees trust them. Only 46% of employees agree. The gap comes from BCG’s second CEO Insomnia Index, which surveyed 645 CEOs at companies with at least $100 million in revenue and 500 nonexecutive employees, and it carries over into how leaders talk about AI.

The AI-specific numbers are sharper. About 40% of employees say their CEO has not been open enough about how AI will affect their roles and careers. Only 15% of CEOs worry that their AI approach could erode trust. Leaders are not weighing a risk their people already feel. The same pattern shows up on well-being: about 70% of CEOs think employees trust them to prioritize it in funding, while 36% of employees agree.

The cost is measurable. According to BCG’s release, half of employees whose trust in their CEO’s AI leadership grew say they will go above and beyond, versus fewer than 10% of those whose trust eroded. Discretionary effort is where an AI rollout either gets adopted or quietly routed around.

This lines up with McKinsey’s August argument that trust, not technology, decides whether an AI transformation captures value. McKinsey found roughly one in five employees report anxiety about AI-driven changes, rising to one in four among middle managers, and that employees with low trust in organizational support are 1.5 times more likely to report that anxiety.

The takeaway for anyone running an AI program — the plan needs a plain statement of what changes for which roles, delivered before the tooling lands. BCG’s Matthias Tauber puts the standard as realistic words and consistent actions. Ship the communication on the same schedule as the software, and measure trust the way you measure adoption, because the 46% figure suggests most CEOs are currently guessing.