AI agents: your data, your rules. Not Silicon Valley's.
Connecting applications and letting an AI act has become easy. Deciding what it can actually reach, at the moment it acts, is much harder. This session presents the architecture implemented at Ariovis: in-house skills, authorization checked at every action and a measurement of the gains achieved.
- Wednesday, October 28, 2026
- 11:00 AM – 12:00 PM (Paris time)
- 1 hour
This replay is in French, with English subtitles available.
Video hosted on YouTube: the player only loads when you start playback.
A connector should not decide your company's rules
The “plug & play” promise is appealing: you connect your applications, the agent acts, tasks move forward. The problem appears one level down. A vendor's native connector implicitly embeds an access model: which data can be reached, which actions are allowed, within which limits. That model was designed for the product, not for your organisation, your processes or your obligations.
At Ariovis, we use these technologies without delegating our rules to them. We develop our own skills to teach the AI how to work with our tools and procedures, and we check its authorizations at the moment of each action, according to our business rules and the context. Knowing how does not grant the right to do: this is the distinction between what an agent can execute and what it is authorized to execute here and now.
This is exactly how we approach digital identity: connecting security, architecture, operations and business needs rather than treating them separately. We have applied to our own organisation an architecture designed for large-enterprise requirements, and we are coming to show what we do rather than present a theory.
What you will take away
Make informed choices about native connectors
Understand what a connector decides for you and when developing your own skills becomes necessary.
Check authorization at the moment of action
See how an agent's rights are evaluated in real time, according to business rules and the context of the request.
Monitor usage
Know what is logged and observed to keep control over what agents actually do.
Measure the investment
Understand the method used at Ariovis to relate automation gains to the costs incurred.
Agenda
- 1
The architecture chosen at Ariovis and our choices regarding native connectors
- 2
Developing our own skills: teaching the AI our tools and procedures
- 3
Demonstration: real-time authorization control
- 4
Usage monitoring and measurement of the gains achieved
Speakers
Matthieu Filizzola
Founder of Ariovis, IAM & PAM expert — Ariovis
An identity and access management expert for more than 10 years, Matthieu helps organisations design and implement secure architectures. His areas of expertise include identity governance, privileged access, fine-grained authorization and securing AI agents.
Quentin Brisé
IAM Consultant, AccuKnox and Axiomatics expert — Ariovis
Quentin Brisé mainly works with AccuKnox and Axiomatics to establish complete visibility and governance over the rights and access of AI systems. His goal is to track, monitor and ensure the compliance of AI actions within our clients' infrastructures.
Practical information
- Date
- Wednesday, October 28, 2026
- Time
- 11:00 AM – 12:00 PM (Paris time)
- Duration
- 1 hour
- Format
- Webinar replay
- Access
- Available online
Further reading
AI agent security
Our service dedicated to governing agents and their access.
Read moreFine-grained authorization
Decide at the moment of action, based on context, rather than granting standing rights.
Read more