Wednesday, 29.04.2026, 17:30 live @ i.sd Foyer

Strange Rooms, or the Ontology of Inference

Architecture has always built strange rooms: the grotto, the folly, the chamber at which the rules of the order begin to buckle. The machine learning model turns this architectural intuition into operational logic, compressing centuries of built form into a latent space from which each inference returns as a strange room of its own. This lecture develops inferentialism as the theoretical register demanded by a machine that learns, working through three SPAN projects in which estrangement operates as architectural method.

Dr. Matias del Campo is an architect and theorist whose research addresses the epistemological implications of artificial intelligence for architecture. His work positions AI as a cultural technique that reconfigures how design disciplines produce and reason about form. Central to his framework are the concepts of inferentialism and the ontology of latent space, developed across publications including Neural Architecture (2022), Diffusions in Architecture (2024), and three guest-edited issues of Architectural Design. He is a recipient of the ACADIA Award. He is co-founder of SPAN and of AKI — Angewandte Künstliche Intelligenz, a Vienna-based company developing AI solutions for industry and public institutions. He teaches at TU Wien, and his work is held in the permanent collections of MAK Vienna, MAXXI Rome, the Albertina, and the FRAC Centre.