C018 — Claim Definition¶


Research	R0055 — RLHF Yes-Men Claims
Run	2026-04-01
Claim	C018

Claim as Received¶

40% of users apply zero scrutiny to AI outputs

Claim as Clarified¶

40% of users apply zero scrutiny to AI outputs

BLUF¶

Partially correct. A Microsoft/CMU CHI 2025 study found participants self-reported using no critical thinking for 40% of their tasks when using AI. The nuance: this is self-reported behavior for a percentage of tasks, not 40% of users applying zero scrutiny to all outputs. The distinction matters.

Scope¶

Domain: AI alignment, sycophancy, enterprise AI
Timeframe: 2022-2026
Testability: Verifiable against published research and documentation

Assessment Summary¶

Probability: Likely (55-80%)

Confidence: Medium

Hypothesis outcome: H2 prevails — see assessment for details.

[Full assessment in assessment.md.]

Status¶

Field	Value
Date created	2026-04-01
Date completed	2026-04-01
Researcher profile	Phillip Moore
Prompt version	Unified Research Methodology v1
Revisit by	2026-10-01
Revisit trigger	Replication with larger sample; observational (not self-reported) measurement of scrutiny