Patrick Butlin is a philosopher of mind and cognitive science who serves as Senior Research Lead at Eleos AI. His work centers on the philosophical underpinnings of artificial intelligence—most notably, whether and how AI systems might possess mental capacities like consciousness and agency that could warrant moral consideration. Butlin explores which cognitive features matter, what it would take for AI to truly have them, and how to rigorously assess such capacities.

Butlin’s scholarship is marked by its sharp focus on the nature of representation, the mental direction of fit in desire, and how these issues bear on both human and artificial cognition. His Ph.D. thesis, The Direction of Fit of Desire (King’s College London, 2015), challenged conventional views by arguing that desires represent values rather than direct instructions for action. Since then, he has published notable work including Reinforcement Learning and Artificial Agency (2024), which investigates the minimal conditions for AI agency through the lens of reinforcement learning; The Agency in Language Agents (2024), exploring whether advanced language-based AI systems truly exhibit agency; and Affective Experience and Evidence for Animal Consciousness (2020), examining what evidence supports conscious experience in nonhuman animals PhilPapers.

Filter

Sort