Two months ago, the Amphipôle at UNIL filled up early. Coffee in hand, researchers from every corner of campus walked in not quite knowing who they’d meet, that was the whole point of AI@UNIL Day.
Liudmila Zavolokina’s keynote, “Slow Down or Speed Up? Designing AI Systems for Human Judgment,” set the tone: a question that felt less like a talk title and more like something everyone was quietly asking themselves.
The other talks jumped from cognitive science to sepsis detection, from genetics to crisis hackathons, proof that “AI research” at UNIL means very different things depending on which door you walk through. The poster session turned the hallway into its own mini-conference, full of pitches and debates over sandwiches. And when the aperitif rolled around, nobody seemed in a hurry to leave.
What stays with us, two months on, is the sense that for one day AI became a language the whole university could speak together.
