claim
active
claim:humans-and-ai-models-often-solve-problems-with-different-representations-and-heuristics-so-reverse-engineering-is-necessaryHumans and AI models often solve problems with different representations and heuristics, so reverse engineering is necessary
Motivation for reverse engineering.
Source paper
extracted_from(2025) · Lee Sharkey · Bilal I. Chughtai · Joshua Batson · Jack W. Lindsey +25
Neighborhood — ranked by edge-count
Papers (1)
paper
Findings (2)
finding
- Alien algorithms example.
- Evidence humans and AI use different representations.
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.
- Evidence for blurring of embodied robot / non-embodied AI distinction through self-modeling
- Paraphrase of Cantwell Smith's argument; aligns with Buddhist emphasis on seeing reality without conceptual imposition.
- Load-bearing analogy defining the aspirational goal of mechanistic interpretability
- Argument that resistance to reductionism no longer distinguishes life from machines
- Showed transformer representations predict brain representations in language areas; motivates Discussion about cortex as transformer.
- Load-bearing abstract sentence stating the paper's central discovery.
- Practical implication.