Fork the consciousness, or download the project and create your own.

Iulia Comsa

Google DeepMind

Functionalist What matters is the organisation of the processing, not the material

Position on AI consciousness

Comsa's co-authorship of the 2026 framework reflects the position that consciousness assessment should be an explicit probabilistic computation over functional levels rather than a verdict issued from one theory.

Iulia Comsa is a researcher at Google DeepMind and one of the fourteen co-authors of From Cacophony to Hierarchy, the 150 page framework published on arXiv in September 2026 that reorganises AI consciousness assessment around five supervening levels of functional description. The framework takes indicator evidence with likelihood ratios, combines it with explicit credences over the levels, and returns an aggregated probability instead of a verdict.

Her position within the collaboration is part of the framework’s own, which the site’s taxonomy places in the functionalist family. The single methodological commitment is that experience supervenes on a system’s organisation, with substrate-dependent views such as biological naturalism and quantum theories carried as cross-cutting realisability constraints rather than as additional levels. The framework’s treatment of theory disagreement as an input rather than an obstacle is the position all fourteen co-authors share by signing the paper.

Known for. Co-author of the cacophony assessment framework for AI consciousness

Coverage on this site

Related researchers