AI and Databases: a hate-love story

AI and Databases: a hate-love story

Quand

20 novembre 2026    
10 h 30 min - 12 h 00 min

Où

Saint Priest Campus – Building 5 – Room 01.124
860 rue St Priest, Montpellier

LIRMM

AI and databases have had a complicated relationship for decades. They need each other, depend on each other, and yet have often behaved as if they belonged to different worlds. Databases provide AI with the data it needs (to learn, reason, answer …), while AI could make databases smarter, by learning from data and queries. But their relationship has also been marked by misunderstandings, competition, and occasional breakups: different communities, different research agendas, and different beliefs about what constitutes a fundamental problem.

The arrival of modern data-driven AI, in particular, LLMs, has attracted much mutual interest. AI is changing the way we query, optimize and integrate data. Databases, meanwhile, are becoming essential for building AI systems that can scale and provide guarantees about data quality, provenance, reproducibility, and trust. Furthermore, new technologies such as vector databases, retrieval-augmented generation (RAG) and AI-assisted data management are creating strong synergy between the two fields.

In this talk, I revisit the history of this “hate-love story,” from early attempts in the 1980’s to bring intelligence into “expert database systems” to today’s AI revolution. I examine what AI is bringing to databases, what databases can bring to AI, and where the two communities still fundamentally disagree. Finally, I discuss opportunities for further research in this area.

Carte non disponible