Shadow AI: Unsanctioned AI Use in the Workplace and How to Respond

'Shadow AI' means employees using AI tools at work that their organisation has not approved, monitored or risk-assessed. It is not a fringe problem: across recent European workplace research it is closer to the norm than the exception.

This workshop is not about blame or prohibition. It is about informed decision-making. We begin with what the evidence actually shows about the scale and nature of unsanctioned AI use, then examine the three strategic responses open to any organisation — ban, tolerate, or formally enable — and the practical consequences of each.

The workshop is structured around a core decision-making exercise: participants work through a set of real Shadow AI scenarios and debate the appropriate organisational response. By the end of the session, each participant has drafted a Shadow AI policy position or a set of governance principles they can take back to their organisation.

Designed for managers, HR professionals, compliance officers, and cybersecurity teams. No technical background is required.

• What Shadow AI is: definition, scale, and why it matters now
• Research findings: what European employees are actually using and why
• Strategic responses: ban, tolerate, or formally enable — consequences of each
• Scenario workshop: working through real Shadow AI situations
• Drafting your response: a Shadow AI policy position or governance principles

By the end of this training, the participant will be able to:

• Define Shadow AI and explain why it represents a governance challenge for organisations
• Describe the three main strategic responses to unsanctioned AI use and the implications of each
• Apply a structured decision framework to real Shadow AI scenarios
• Draft a Shadow AI policy position or set of governance principles for their organisation

Opens with data-driven trainer input, then moves immediately into group scenario exercises. The final session is a structured drafting exercise producing a Shadow AI policy position. Discussion-heavy and participant-led.