Workers are adopting artificial intelligence in the workplace faster than many employers. Employees are increasingly turning to AI for their daily tasks, but businesses and organizations have been slower to develop policies or initiatives that actively encourage its use, according to an analysis published by the economic journal ESB.
Across every sector, more employees report using AI tools such as ChatGPT and Copilot than say their employers actively encourage or support their use. The gap exceeds 15 percentage points in most industries and approaches 30 percentage points in the public sector and education.
A survey conducted by De Nederlandsche Bank (DNB) last October found that only 25 percent of employees feel their employers actively support the use of AI at work. Many others said they are permitted to use certain AI tools but receive little encouragement to do so, while some said their employers have no clear policy governing AI use.
Because official company rollouts lag heavily behind individual usage, employees are widely adopting a habit known as “Bring Your Own AI” (BYOAI). They log into consumer-grade versions of ChatGPT or Claude using personal accounts on corporate networks.
This behavior creates a severe governance blind spot. Workers frequently feed proprietary corporate data, client information, or sensitive source code into public models without realizing they are forfeiting data privacy.
The reason the adoption gap approaches 30 percentage points in education and government is rooted in strict bureaucratic regulations and data protection laws (such as GDPR).
While teachers and civil servants use generative tools to rapidly draft lessons, summaries, or reports, institutional policies remain paralyzed by legal uncertainties regarding copyright, accountability, and student/citizen data safety.
In contrast, sectors like IT, finance, and professional services show much smaller gaps. Companies in these fields are more aggressively purchasing enterprise-grade, closed-loop AI software licenses like Copilot for Microsoft 365 to actively capture immediate billable productivity gains.
Recent studies reveal that while over two-thirds of workers using generative AI save anywhere from three to eight hours per week, it rarely results in a lighter workload.
Instead of working fewer hours, the lack of structured employer guidance means the saved time is immediately filled with a higher volume of tasks. This compresses deadlines, increases corporate speed expectations, and inadvertently leads to workers feeling more rushed and overwhelmed than they did before adopting AI.