AI BriefingBuilding RAG-Based Data Sovereignty at a French Public Institution
Key point
France's Banque des Territoires built a RAG solution that ensures data sovereignty.
Details
France's Banque des Territoires (BdT) developed a RAG (Retrieval-Augmented Generation)-based data solution for the EduRénov program, which supports energy retrofits of public school facilities.
The project was carried out in collaboration with Polyconseil and Hugging Face, with the main goal of optimizing the repetitive question-and-answer process between experts and local communities by leveraging a vast amount of specialized documents.
Since the public sector handles particularly sensitive data, the project adopted open-source models and a Sovereign Data Solution instead of commercial APIs such as ChatGPT, securing both data security and strategic sovereignty.
Through this, the plan is to support 10,000 school facility retrofit projects and maximize operational efficiency to achieve energy-saving targets.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Qwen3.5-based Small Decision Model 'Kev' Released with TypeSafe API Compatibility and Performance Metrics