GPT, insofar as it is inner-aligned, is a simulator which can simulate agentic and non-agentic simulacra.

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

• GPT's ability to simulate text automata is the source of its most surprising and pivotal implications for paths to superintelligence.claim0.827
  Importance of recursive generation.
• GPT is behavior cloning, but it is the behavior of a universe that is cloned, not of a single demonstrator.claim0.817
  Broadening behavior cloning to universal simulation.
• GPT doesn't seem to care which agent it simulates, nor if the scene ends and the agent is effectively destroyed.claim0.812
  Illustrates the simulator-simulacra distinction.
• GPT is corrigible in a negative sense because the agent specification (prompt) is not fixed by the policy and the policy lacks direct training incentives to control its prompt.claim0.811
  GPT's corrigibility explained.
• What we call GPT's 'downstream behavior' is the behavior of simulacra; it is primarily through simulacra that GPT has potential to perform meaningful work.claim0.802
  Clarifies where agency resides.
• Treating GPT as an unsupervised implementation of a supervised learner leads to systematic underestimation of capabilities.claim0.798
  Critique of the oracle/supervised frame.
• I do not think any simple modification of the concept of an agent captures GPT's natural category; GPT is not a roleplayer, only that it roleplays.claim0.792
  Rejection of the agent interpretation.
• GPT does not generate rollouts during training, so there is no reason to expect that GPT will form preferences over the consequences of its output related to the text prediction objective.claim0.773
  Argues against instrumental convergence in GPT.