Alizé at the Data Summit Luxembourg 2025: Transparency at the Core of Trustworthy AI

Alizé recently took part in the Data Summit Luxembourg, organized by the Luxembourg National Data Service (LNDS), where our Co-founder and CTO Luis Pérez was invited to speak on a panel hosted by the Luxembourg AI Factory.

During his intervention, Luis addressed a critical and often misunderstood aspect of modern AI systems: large language models are engineered to sound confident and correct, even when they do not truly know the answer. This characteristic makes them powerful tools, but also potentially convincing liars if not designed and used responsibly.

This observation directly reflects one of the core principles behind Alizé’s architecture. From the very beginning, Alizé has been built with a strong focus on transparency and traceability. Rather than relying on generic or opaque model outputs, our system feeds the AI with verified, structured legal data from our own databases. At the same time, Alizé programmatically includes direct references to the original legal sources, allowing users to verify and trace every assertion made by the AI.

When asked what advice he would give to anyone building—or considering building—an AI company, Luis was clear and pragmatic:
focus on added value. An AI product must solve a real problem. It should save time, reduce costs, or meaningfully improve some aspect of the user’s work or daily life. Without that, even the most advanced technology remains just a technical curiosity.

We are proud to see Alizé represented at events like the Data Summit Luxembourg and to contribute to the ongoing discussion around responsible, transparent, and value-driven AI in Europe.