The question of whether and under what conditions machines could be conscious has shifted from a marginal philosophical concern to an active research program. Existing frameworks enumerate necessary conditions for consciousness as flat lists, treating consciousness as a single explanandum addressed by a single theory. This paper argues that work on conditions for consciousness is better organized around a three-facet taxonomy that distinguishes conscious experience as the broad foundation encompassing all phenomena registered within consciousness, conscious awareness as the focused, attentional subset that filters and prioritizes content, and subjective experience as the qualitative, personal dimension characterized by qualia. I defend the distinctness of these three facets against the strongest objections, organize ten conditions across them drawing on a developed account from Etukuru (2025), and compare the resulting framework against Integrated Information Theory (IIT), Global Workspace Theory (GWT), and Higher-Order Theories (HOT). The three-facet structure yields three results. (i) Each major existing theory is shown to address primarily one facet. (ii) Current AI systems are shown to make uneven progress across the three facets, with subjective experience essentially untouched. (iii) Verification asymmetries across facets are shown to have direct implications for AI moral-status debates. The paper brackets the question of how these conditions could actually be realized in a machine substrate. That mechanism question is addressed in a companion paper.
Raghurami Reddy Etukuru (Wed,) studied this question.