Bender et al. 2021: On the dangers of stochastic parrots

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

• stochastic parrotconcept0.766
  Derogatory term for LLMs; Nix's commentary frames it as camp opposite to nascent consciousness.
• The risk of under-attribution of AI welfare appears to be both reasonably likely and reasonably harmful.claim0.710
  Key motivation for precautionary action.
• The adaptation must be specific, not a stochastic trick in which random variation creates an illusion of variety.claim0.701
  Distinguishes genuine context-driven adaptation from mere statistical randomness.
• The adaptation must be specific. We are not talking about a stochastic trick in which random variation creates an illusion of variety and repetition.quote0.701
  Key distinction between genuine living structure and pseudo-random variation.
• Verbal reports, homology, and phylogenetic provenance are insufficient to determine sentience in unconventional agents.claim0.700
  Core claim: Turing test and brain homology fail for synthetic, AI, and radically non-human agents; new frameworks required.
• Sentience assessment should seek deep invariants across possible minds, not arbitrary criteria tied to evolution on Earthclaim0.698
  Core normative claim: frameworks must identify fundamental properties of sentience independent of phylogenetic accident or familiar substrates.
• Undue anthropomorphism is surely detrimental to the public conversation on AIclaim0.697
  Normative motivation for the paper's conceptual contribution
• AI Control: Improving Safety Despite Intentional Subversion (Greenblatt et al. 2024)concept0.696
  Related work studying capability of LLMs to subvert safety measures if severely misaligned