AI agents no longer just generate content – they make decisions and take action. So who governs them, and how? This topic brought our 100 Women in Finance Paris community together on 1 October for a session on “Governing AI From Day One: Rethinking AI Trust in Financial Services”. The starting point of our discussion was that as AI agents move across data, systems and business processes, governance can no longer be an afterthought.
Moderated by Renan Borne, Linedata, the conversation brought together Susan Haimet, Saavvy AI and Olivier Poupeney, Linux Foundation, FINOS who shared their insights and perspectives.
What stayed with us:
- How Governance-as-Code can support greater agent autonomy while preserving control, accountability and compliance
- How to translate business policies and risk requirements into enforceable rules for autonomous agents
- Where the line sits between human approval, intervention and full autonomy
- A shared conviction: governance and regulation for AI have to be built collectively, before autonomy outpaces oversight.
Thank you to Linedata for hosting the session and their warm welcome and to our speakers for their contributions. The cocktail turned into a second session of its own – and was a good reminder of how much there is still to build together.
