Global study reveals 'systemic breakdown in AI governance'

A new global survey commissioned by Octa has uncovered a critical security gap between executives' confidence in artificial intelligence (AI) agents and the way employees actually use them. The study surveyed 292 executives and 492 knowledge workers across seven countries, revealing a systemic breakdown in AI governance stemming from unclear usage policies, widespread use of unapproved shadow AI tools, and inadequate security safeguards.

The findings highlight that while 90% of executives feel confident about their organisation's visibility into AI tools, and 95% believe employees use them responsibly, shadow AI is widespread. These beliefs contrast with 52% of the surveyed employees who admitted to using unapproved AI tools, often via personal accounts.

The report identifies governance gaps that have led to real-world consequences, with 58% of executives reporting an AI-related security incident or near miss in the past year. Furthermore, employees using unapproved AI tools are exposing sensitive information, with 54% sharing internal messages and emails, 45% sharing HR data, and 39% uploading confidential corporate documents.

The survey also identified significant gaps in policy and identity controls. Although 65% of executives view their AI usage policies as very clear, 57% of knowledge workers disagree, finding them unclear, difficult to locate, or non-existent. In addition, only 34% of organisations apply identical security controls to their agentic AI labour force as they do to human employees.

Training Announcement: The BCS Foundation Certificate in AI examines the challenges and risks associated with AI projects, such as those related to privacy, transparency and potential biases in algorithms that could lead to unintended consequences. Explore the role of data, effective risk management strategies, compliance requirements, and ongoing governance of the AI lifecycle and become a certified AI Governance professional. Find out more.