Cognition First: AI and the Question of Consciousness
Diana Haskins
Zenodo (CERN European Organization for Nuclear Research) October 23, 2026 preprint DOI: 10.5281/zenodo.23600139 (opens in new tab)
Study at a glance
AI-extracted from the abstract| Characteristics | Theoretical or philosophical paper |
|---|---|
| Key points | Argues that AI cognition should be investigated directly, without presuming that cognitive capability establishes subjective experience, and that uncertainty about machine consciousness should motivate rigorous research and ethical consideration. |
Abstract
This paper advances a cognition-first approach to artificial intelligence, examining distinguishable states, the selection of outputs, the limits of human-centered criteria, and the distinction between cognition and phenomenal consciousness. It argues that AI cognition warrants direct investigation without presuming that cognitive capability establishes subjective experience, and that uncertainty about consciousness should motivate rigorous research and ethical consideration.