Who Is Liable When an AI Agent Makes a Business Decision?

Growing regulatory scrutiny of artificial intelligence is raising questions about accountability when AI agents access business systems, process personal data or take actions autonomously. In October 2026, the UK Information Commissioner’s Office expanded its scrutiny of AI developers and sought evidence on the data-protection risks associated with agentic AI. Businesses adopting these technologies must consider access controls, data protection, human oversight and liability for unauthorised actions. As AI adoption accelerates, organisations need governance frameworks that establish clear accountability before deploying autonomous systems.