The framing that rogue employees circumvent AI policies misses where the biggest exposure actually sits. Senior leaders, with elevated access to sensitive systems, administrative privileges, and a broader view of strategic data, are disproportionately the ones bypassing controls. TrustedTech's research finds that 51% of those same leaders say they are concerned about staff using unapproved AI — a concern they are not applying to themselves.
The motivations behind shadow AI adoption reveal something more structural than individual carelessness. Among those using unapproved tools, 24% cite limited access to employer-approved alternatives, 21% say the unapproved tools simply work better, and a further 21% do not want their usage data accessible to their employer. Cost, productivity pressure, and a quiet mistrust of internal monitoring are driving the behaviour from the top down.
That monitoring anxiety runs in both directions. Twenty-eight percent of UK employees say they are worried their employer tracks AI usage; 21% judge colleagues who rely on AI heavily; and 23% actively reduce their own AI use to avoid scrutiny. The result is a workplace where AI adoption happens in parallel — visible and sanctioned on one layer, informal and ungoverned on another.
The numbers put a specific shape on the compliance risk. Seventy-six percent of UK employees acknowledge that unapproved tools carry security or data privacy risks. Nearly half (47%) use them anyway. When the people responsible for setting and enforcing governance policy are themselves the primary source of the behaviour they are supposed to police, the usual remedies — training programmes, acceptable-use policies, monitoring dashboards — address the wrong layer of the organisation.
Julian Hamood, Founder and Chief Visionary Officer of TrustedTech, said: "Businesses often assume Shadow AI is a bottom-up problem, but our research shows it is being driven from the top down. Senior leaders are not only the biggest users of unapproved AI tools, they are also knowingly bypassing safeguards because the perceived benefits outweigh the risks. When that behaviour is modelled at the top of an organisation, it becomes significantly harder to enforce governance elsewhere in the business.
TrustedTech is a Microsoft Cloud Solutions Provider. The full whitepaper, "Shadow AI in the Workplace," is available at trustedtechteam.com.
