How do algorithms work? As algorithmic systems—from Google’s search engine to Spotify’s music recommender—have become objects of popular concern, this question has proven vexing. With the advent of techniques like deep learning, algorithmic systems are often described as “uninterpretable”—so complex that it is impossible, even for insider experts, to explain their outputs. And yet, engineers, like ordinary users, are tenacious interpreters, eager to make sense of algorithmic behavior, regardless of its internal complexity. This talk draws on ethnographic fieldwork with developers of algorithmic music recommenders in the US to theorize “interpretability,” describing how engineers interpret supposedly uninterpretable systems. Theories of acousmatic listening offer useful models for making sense of this interpretive work, for the engineers as well as outside critics.

Bio: Nick Seaver is Associate Professor of Anthropology at Tufts University, where he also directs the program in Science, Technology & Society. His research has appeared in venues including Cultural Anthropology, Cultural Studies, Big Data & Society, and the Journal of the Royal Anthropological Institute. He is the author of Computing Taste: Algorithms and the Makers of Music Recommendation (2022), an ethnographic study of music recommender system developers. His current research examines the rise of attention as a value and virtue in machine learning worlds.