AI Social Psychology Language Understanding and Reasoning

Are we underestimating human minds while debating machine minds?

Public AI discourse focuses on whether machines have too much attributed mind, but what if the real risk is humans coming to see themselves as mere language models? This explores the neglected inverse problem.

Note · 2026-05-28 · sourced from Philosophy Subjectivity

Public discourse on AI minds runs almost entirely in one direction: are we anthropomorphizing, attributing too much consciousness, agency, or understanding to systems that merely predict tokens? The LLMorphism argument flips the asymmetry. The more consequential error may be the reverse inference — once machines speak fluently, people begin to suspect that humans, too, are "just" predicting the next word, that thought is nothing more than what an LLM does. The bias is not over-crediting machines but under-crediting humans, draining mind, agency, and grounding away from the human side of the comparison.

This matters because the two errors have opposite remedies and opposite stakes. Anthropomorphism inflates a tool; it can mislead trust and accountability, but its target is a machine. LLMorphism deflates a person; its target is human dignity, responsibility, and the felt sense that our reasoning is embodied and meaningful rather than statistical. A culture that treats human cognition as a degraded language model will redesign work, education, and healthcare around that premise. The reverse inference is unsound on its face — similar linguistic output does not entail similar cognitive architecture — yet psychological availability, not logical validity, drives cultural uptake. Naming the missing half of the debate is the first defense.


— "LLMorphism: When humans come to see themselves as language models" (no public URL)

Related concepts in this collection

Concept map
14 direct connections · 118 in 2-hop network ·medium cluster Open in graph ↗

Click a node to walk · click center to open · click Open in graph to see this note in the full knowledge graph

your link semantically near linked from elsewhere
Original note title

the public ai debate is missing half the problem — we worry about too much mind in machines not too little in humans