Dezhi Luo
.jpg?apiToken=eyJhbGciOiJIUzI1NiIsImtpZCI6IjIwMjMtMDUtMDQifQ.eyJzY29wZSI6Im5vdGUtZXhwb3J0IiwibmlkIjoiYllPd0tQZzlCcmZfamciLCJpYXQiOjE3ODk0NDA1MzQsImlzcyI6Imh0dHBzOi8vc2xpdGUuY29tIiwianRpIjoicjA1ZjVvVDZmb3JhOXciLCJleHAiOjE3OTIwMzI1MzR9.zl8wKo9srW1ZJJCWFWXqMoXNsd8ez2obC0X-Inio_gs)
United Kingdom
PhD Student
University College London (UCL)
Machine Learning/Technical AI, Neuroscience/Cognitive Science, Philosophy/Ethics, Interdisciplinary
United Kingdom PhD Student University College London (UCL) Machine Learning/Technical AI, Neuroscience/Cognitive Science, Philosophy/Ethics, Interdisciplinary
Project
(The philosophical foundations for) a formal toy framework for implementing the iterative natural kinds strategy in consciousness science.
Project Overview
The Iterative Natural Kinds (INK) strategy (Bayne et al., 2024) offers a principled basis for extending consciousness tests beyond humans by grounding attributions in population similarity rather than intuition or specific test outcomes. While influential, the strategy remains largely conceptual, lacking formal specification of how populations, evidence, and similarity relations should be represented and updated. This project develops the theoretical and formal foundations necessary to make the INK strategy tractable as a framework for assessing consciousness in large-scale foundation models. It addresses two interconnected challenges: first, establishing the ontological and epistemic commitments that render scientific tests of consciousness in AI systems defensible; and second, specifying the granularity and structural relations among proximity measures that would enable principled, belief-updating-based extensions across populations. The resulting framework is intended as a foundational tool for future empirical and theoretical work on AI sentience, with direct implications for how consciousness benchmarks should be designed, validated, and interpreted across biological and artificial systems.
Mentor
Collaborators: Pietro Amerio (UCL), Matilda Gibbons (UPenn)
About the Scholar
I am an incoming MPhil/PhD student in philosophy at UCL. Before I was an undergrad at the University of Michigan’s Weinberg institute for Cognitive Science, a visiting student at UCL Faculty of Brain Sciences affiliated with the MetaLab, and an AI Fellow at the Institute of Philosophy. I spent my first-year in the Cognitive Science MA programme at The University of Edinburgh. I am most interested in the theoretical foundations of cognitive science and its implications to the responsible development of AI, with a focus on consciousness, agency, and understanding. Outside of these I enjoy alternative music, post-modern fictions, and random streetwalking.
Links
