Thank you for joining us
Thank you for being part of the conversation. Whether you joined us in Paris or are catching up afterwards, we’ve brought together the ideas, conversations and practical examples from Intelligent Enterprise: Live in Paris.
The central message of the evening was clear: when building costs next to nothing, scarcity moves. An AI agent can now be prototyped in an afternoon, and your competitors have access to the same models, platforms and tools. What remains hard, and what sets organisations apart, is choosing which prototypes deserve to become real uses, knowing whether an agent is right, and redesigning the way work is organised around it. The gap is visible in the numbers: only 13% of CIOs report significant or transformational value from AI (Gartner, October 2026).
In their keynote, Stéphane Beringuel and Pascal Malotti framed the path from proof of concept to MVP around three decisions: choosing, judging and absorbing. Choosing means running a portfolio rather than collecting agents, with a value hypothesis for each use case, an owner who can make the call and a date to extend, redirect or stop. Judging means testing for accuracy, compliance and uncertainty, and keeping those tests when the model changes. Absorbing means changing the work, not just speeding up the task: the large-scale gains from electricity came with the reorganisation of factories, not just with new motors. Their conclusion: responsibility can’t be bought. Someone has to decide, someone has to accept the use, someone has to answer for the result.
The executive panel brought together Marie Jelu (Air Liquide) and Ilias Charki (Louis Vuitton) around one idea: AI first transforms what the customer doesn’t see. Both started with internal uses, where teams’ work changes before customers notice anything. The discussion covered what has to be put in order before starting (processes, data, who has the authority to decide between countries), how to give teams autonomy without losing control of usage and cost, when a use case is mature enough to deploy while technology keeps moving, and how hard it is to turn the time savings promised in a business case into real value on the ground.
In the fireside chat, Arnaud Girard, Head of AI GTM at Google Cloud, brought the technology perspective: who is shaping the AI market, where value is really being created and how leaders can make sound choices in a market that never stops changing. The conversation looked at what organisations should buy as a foundation and what they should keep and build as their own advantage, and at what leaders need to know about every agent (its identity, its rights, its actions) before letting it act on their systems.
The evening continued with drinks, networking and live demos.
