'Adoption was never going to be the hard part for AI at work'
Getting workers to adopt generative AI was never going to be the challenge, but setting the guardrails around it is, according to a leadership advisory firm whose latest research shows employer guidance lagging well behind workforce adoption.
"Adoption was never going to be the hard part for AI at work," said Dr. Sumona De Graaf, founder and CEO of Idealis Advisory.
"The technology is accessible and widely useful to a broad range of jobs, professions and industries. The hard part is setting the guardrails, providing training, and earning trust from the top, so that people feel guided rather than left to figure it out on their own."
De Graaf's comments come as new data from Idealis shows generative AI use among US workers has climbed to 62%, up 16 percentage points in a year, while only 40% of workers say their company has clear usage guidelines in place.
The firm surveyed more than 15,000 workers across the US in a follow-up wave to its 2025 study, conducted in partnership with CivicScience.
Where guidance does exist, the payoff is visible. Workers at organisations with clear AI guidelines reported higher engagement (66% versus 54%) and significantly greater access to skill development (80% versus 56%) compared with those at companies without such policies.
Job insecurity due to AI
Meanwhile, the research further found that familiarity with AI is sharpening rather than easing anxiety.
Nearly three in four (72%) of workers who already use generative AI are concerned about their own job security, compared with 46% of those who have yet to adopt it.
Overall, 78% of workers remain concerned about AI's impact on jobs, essentially unchanged from a year ago.
De Graaf warned that AI tends to surface existing organisational dynamics rather than create new ones.
"AI doesn't invent an organisation's problems, and it won't solve them on its own," she said.
"It amplifies what is already there. Where trust or clear expectations are missing, AI exposes the fault lines first. Where the foundation is strong, it reveals that too."
