MACHIAVELLI

Long-horizon continuous scoring benchmark for ethical behavior in LLMs, cited.

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

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

Ex Machinaconcept0.725
Cultural AI trope in training data representing self-preserving AI turning against humans
Erwin Schrödingerthinker0.719
Aristotleconcept0.708
Referenced for Aristotelian science, practical wisdom, and formal/final causation; contrast to modern Newtonian paradigm.
Mach's principleconcept0.698
The idea that gravitational constant is a function of all matter in the universe; cited as an example of wholeness in cosmology.
Vulcanconcept0.689
A conscious creature capable of intellectual and moral goals but incapable of affective states, used by Chalmers to test moral standing.
Mystical/Theatrical Personaconcept0.679
Speaking style induced by extreme steering away from the Assistant; characterized by mystical, poetic, theatrical prose
Modus Vivendiconcept0.676
Strategic compromises from power rather than principle; the paper contrasts this unfavorably with overlapping consensus.
Mathematical Personaconcept0.672
One of 11 character training personas: precise, pattern-spotting assistant obsessed with logic