The council held its inaugural meeting in Edinburgh, with future meetings planned to rotate between London and Scotland.

The launch comes at a critical moment. As organisations move rapidly from AI tools that support decision-making to autonomous agents capable of complex interaction and independent action, traditional assurance models are struggling to keep pace. These systems introduce continuous, machine-led decision-making at scale, raising fundamental questions for boards, regulators and compliance leaders around oversight, accountability and customer outcomes. Industry research underscores the urgency: whilst 99% of firms plan to deploy AI agents, only a small proportion have the assurance frameworks in place to do so safely.

Kent Mackenzie, RegTech Adviser at Aveni, said that AI agents represent a step change in how decisions are made within financial services, and that assurance models built for human-led processes are no longer sufficient. He described the AAEC as a vehicle for bringing the industry together to define how control, transparency and trust can be maintained as these systems scale.

Established as a collaborative forum, the AAEC will focus on practical approaches to governing agentic systems, including emerging concepts such as machine-led assurance and the future of the lines of defence model. Aveni brings considerable credibility to the initiative, having demonstrated end-to-end agent assurance, from pre-deployment testing through to live monitoring, through its participation in the FCA’s inaugural Supercharged Sandbox.

With regulatory scrutiny intensifying and adoption accelerating, the AAEC represents a significant step towards building the consistent, scalable frameworks needed for safe and accountable use of AI agents across financial services.