AIEN Research · Brief

The Turing: A Unit of Verified Explanatory Gain

The three-page plain-language brief: what the Turing measures, what it does not, and what would prove it wrong.

M. Drake Stapleton · October 2, 2026

v1.43 pagesNot peer reviewed

Abstract

The Turing (T) is a unit of verified explanatory gain: one Turing is one bit of net held-out description-length improvement against a declared baseline, after paying for the explanation itself, under a frozen measurement profile.

This brief states what the unit measures, what it explicitly does not measure, and what would prove the interpretation wrong, in plain language. The full definitions, proofs, and falsification criteria live in the formal paper.

Brief; not peer reviewed.

Research artifacts

Paper (PDF)

SHA-256: 7689932e8ee60690277d3c933b47385a05a11cb69d0985daf3837d18987e3146

paper.pdf

LaTeX source

Complete compilable source.

SHA-256: 4bada1db5740fb8e8bc767fd8838a15d25efab76154f7c86a006eda77667e413

paper.tex

Digest manifest

SHA-256 manifest for this release.

SHA256SUMS

Formal paper

Computing Machinery and Understanding: From the Imitation Game to the Turing (v1.13).

formal paper page

Cite

@misc{stapleton2026turingbrief,
  author = {M. Drake Stapleton},
  title = {The Turing: A Unit of Verified
           Explanatory Gain},
  year = {2026},
  month = {October},
  version = {1.4},
  note = {Brief, not peer reviewed},
  url = {https://aienos.com/research/the-turing-brief}
}