Nearly half of senior AI executives admit their organizations have bypassed formal AI governance policies for urgent deployments, even as 91% now use agentic AI without updated frameworks. This widespread practice creates significant, unmitigated risks, allowing advanced AI systems to operate with insufficient oversight and raising critical questions about accountability.

A stark contradiction emerges: 98% of senior AI executives report having formal AI governance policies, yet 47% admit to ignoring them for urgent deployments. Furthermore, 49% of those deploying agentic AI have not updated their frameworks for these advanced systems. This disconnect between policy and practice creates a substantial governance gap.

Companies are trading immediate speed for long-term control and compliance. This gamble will likely lead to significant regulatory challenges and potential liabilities as governments accelerate their oversight. Prioritizing rapid deployment over safeguards sets a dangerous precedent for global AI regulation.

The Governance Illusion: Policies on Paper, Risks in Practice

  • 98% — of senior AI executives report having formal AI governance policies in place, according to EY (2026).
  • 47% — of senior AI executives admit their organization has previously not followed its AI governance process for urgent deployments, according to EY.
  • 91% — of senior AI executives report their organization uses agentic AI, according to EY.
  • 49% — of respondents whose organization uses agentic AI say their organization’s existing governance framework has not yet been updated to specifically include agentic AI requirements and risks, according to EY.

Despite a high reported rate of formal AI governance policies, actual implementation and adaptation for advanced systems like agentic AI remain significantly lacking. Policies exist on paper, but nearly half become ceremonial when speed is prioritized. This creates a critical gap between stated intent and actual practice. Companies deploying agentic AI operate with a dangerous blind spot: nearly half (49% per EY) haven't updated their governance frameworks for these advanced systems. They are effectively running high-stakes experiments without a safety net, inviting unforeseen consequences.