LABS
Research groups working inside the Center.
Research groups working inside the Center.
Lab
QUALIA Lab studies interpretability, consciousness and alignment in humans and AI systems. We ask what a model’s self-reports and internal states can tell us about whether it has experiences, and how to check that its goals match ours. We emphasise the humanities approach, and see these systems as a way of learning more about ourselves along the way.
What consciousness is in humans, and which of its features could appear in artificial systems.
How humans and AI systems report on their own states, and when those reports count as evidence.
Reading the internal states of AI systems with the standards cognitive science applies to human minds.
What it would mean for an AI system to share human goals, and how anyone could tell.
Lab lead
Assistant Professor, Center for Cognitive Science, University of Zagreb
Machine Learning Lead, Meedan
Research associate
Associate Professor, Faculty of Croatian Studies, University of Zagreb
External research associate
TU Wien
Book · Springer Nature, 2025
Kristina Šekrst
Whether an artificial system can experience the world, followed through philosophy, neuroscience and the architecture of current AI — from cybernetics and the imitation game to large language models and the hard problem.
BookSpringer, 2026
Second edition, from neural networks to large language models and agentic AI.
Sandro Skansi & Kristina Šekrst
Paperin review
Uses mechanistic interpretability to test Dennett’s four marks of qualia against language models. Privacy and ineffability are present, intrinsicality holds only in a weakened diachronic form, and direct apprehensibility is absent.
Kristina Šekrst
Paperwork in progress
Argues that Russell’s Human Compatible framework fails structurally, because training turns the properties that define a beneficial machine into optimisation targets with no way to verify them.
Barbara Geld & Kristina Šekrst
SymposiumALIGN27 proposal, work in progress
On the philosophical and ethical assumptions behind AI alignment and interpretability methods, asking what these methods actually establish and who answers for their failures.
Kristina Šekrst, John Dorsch (Prague), Barbara Geld (Vienna), Scott Hale (Oxford), Erik Persson (Lund), Mirko Daniel Garasic (Rome) and Roman Krzanowski (Kraków), co-organised with Meedan
Everything the Center has published, newest year first, with links to each piece.
OpenKeynotes, conferences the Center partners in, and selected talks by our staff.
OpenPreprints, in-press work and the full talk schedule, kept by the lab itself.
VisitCollaborations, data requests, and invitations to explain why a model saying it is conscious is not yet evidence that it is.
