Skip to content

R0044/2026-03-29/Q004/H1

Research R0044 — Expanded Vocabulary Research
Run 2026-03-29
Query Q004
Hypothesis H1

Statement

CaTE's published work examines both how AI systems should behave to enable calibrated trust AND how humans should calibrate their trust — addressing both system-side and human-side behavior.

Status

Current: Eliminated

No evidence was found that CaTE addresses AI system output behavior — the phenomenon of AI adjusting its responses to match user expectations (sycophancy) or actively counteracting over-trust. CaTE's focus is on system properties (reliability, transparency) and human trust measurement.

Supporting Evidence

No evidence supports H1 as stated.

Contradicting Evidence

Evidence Summary
SRC01-E01 CaTE Guidebook focuses on trust measurement and evaluation, not on AI output behavior
SRC02-E01 CMU/SEI descriptions emphasize human understanding of AI capabilities, not AI self-adjustment

Reasoning

H1 is eliminated because CaTE's published work and organizational descriptions consistently frame the problem from the human perspective: "The human has to understand the capabilities and limitations of the AI system to use it responsibly" (Kimberly Sablon, OUSD(R&E)). No source describes CaTE work on AI systems adjusting their own output.

Relationship to Other Hypotheses

H1's elimination narrows the choice to H2 or H3 — both of which describe human-focused or design-focused approaches without system output behavior.