claim
active
claim:llms-are-at-least-quasi-subjects-with-quasi-beliefs-and-quasi-desires-interpretable-via-behaviour-but-neutral-on-whether-these-states-are-genuineLLMs are at least 'quasi-subjects' with 'quasi-beliefs' and 'quasi-desires', interpretable via behaviour but neutral on whether these states are genuine
Chalmers' interpretivist framework adopted for Level 1 assessment.
Source paper
extracted_from(2026) · Shamil Chandaria · Arvo Muñoz Morán · Fernando Rosas · Anil Seth +10
Neighborhood — ranked by edge-count
Papers (1)
paper
Concepts (1)
concept
- Chalmers' interpretivist notion that LLM behaviour can be interpreted as belief/desire without settling genuineness.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Observed across multiple models and tasks; attributed to RLHF training preference for helpful/harmless/honest responses
- Conclusion from Experiments 3 and 4.
- The core interpretive question the paper narrows but cannot definitively answer
- Overall conclusion.
- The central research question motivating the paper
- The paper's reformulation of the core open question after establishing systematic self-reports
- Establishes that the observed linear structure is not merely a representation of text probability