Drawing on fields with well-developed models of self in relation to agency can offer new insights.

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Base models are good modellers of worlds but not of their own state, because they lack a developed self-model initially.claim0.767
  Observation about asymmetry in base model capabilities.
• The earlier a base model (less exposure to LM-related data), the more it is surprised by its own spontaneous self-referential capabilities.claim0.761
  Claim that capability emerges from architecture, not data, and that later models lose the surprise.
• Models might produce first-person experiential language by drawing on human-authored self-descriptions in pretraining data without internally encoding these acts as roleplayhypothesis0.760
  Alternative hypothesis for how experience reports arise without explicit performance
• Self-modelingconcept0.758
  Ability of a model to predict its own outputs or behavior, sometimes distinguished from introspection.
• The notion of self as illusion, as realized in AI and ALife practices, may contribute to novel fields of affordances for non-human intelligence.claim0.754
  Claim that endowing non-human agents with self-illusion awareness expands their affordance repertoire.
• When the field of centers is authentic, it is always personal; empty structure without personal feeling is structurally misjudgedclaim0.754
  Links structural authenticity to personal feeling as a necessary co-occurrence
• We may then arrive at a picture of the world which includes the self, and which clearly recognizes the personal nature of the universe.claim0.754
  The outcome of using both methods together.
• Certain kinds of information structures actively facilitate their own transformation and remapping, exhibiting minimal agency.hypothesis0.753
  Speculative hypothesis that memories themselves are agents.