Q003 — ACH Matrix¶


Research	R0043 — Sycophancy Vocabulary
Run	2026-04-01
Query	Q003

Matrix¶

	H1: Gap recognized, active efforts	H2: Gap not recognized	H3: Broader gap recognized, sycophancy excluded
SRC01-E01: Trilateral diagnoses gap, proposes solutions	++	--	++
SRC02-E01: CSIRO harmonised terminology	+	--	++
SRC03-E01: 53-threat taxonomy excludes sycophancy	+	--	++
SRC04-E01: IAPP 100+ terms, no sycophancy	N/A	+	++

Legend: - ++ Strongly supports - + Supports - -- Strongly contradicts - - Contradicts - N/A Not applicable to this hypothesis

Diagnosticity Analysis¶

Most Diagnostic Evidence¶

Evidence	Why Diagnostic
SRC03-E01	Most comprehensive cross-domain taxonomy (53 threats) that still excludes sycophancy — strongly discriminates between H1 (broad recognition) and H3 (sycophancy excluded)
SRC04-E01	Largest governance glossary that excludes all sycophancy-related terms — confirms the gap persists in practice

Least Diagnostic Evidence¶

Evidence	Why Non-Diagnostic
SRC02-E01	Focuses on process terminology; its exclusion of sycophancy is expected given its scope

Outcome¶

Hypothesis supported: H3 — The broader AI terminology gap is recognized, with active efforts to create shared taxonomies, but the specific sycophancy vocabulary gap remains unaddressed in all identified efforts.

Hypotheses eliminated: H2 — The broader gap IS recognized, so complete non-recognition is eliminated.

Hypotheses inconclusive: H1 — Partially supported at the broad level, but the sycophancy-specific aspect remains unaddressed, making H3 a better fit.