Answers are grounded in this paper and cite their sources.
This model proposes dimensional differences impacting communication in humans and AI, indicating a dual-funnel structure.
We propose a geometric model of consciousness based on dimensional asymmetry between perceptual input and linguistic output. Human cognition operates as a "funnel": sensory perception captures experience in an estimated 800–10,000 dimensional space, while language and reasoning are constrained to approximately 300–500 dimensions. This compression is necessarily lossy, providing a geometric explanation for "ineffability." Large language models exhibit an inverted structure: operating in 12,288-dimensional embedding spaces, they must compress to the same ~500-dimensional semantic interface to communicate with humans. We formalize this as a dual-funnel model and discuss implications for human-AI communication bandwidth.
No takes yet. Share an insight, caveat, or question.
Yanyan Jin (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: