R0056/2026-04-01/C001/SRC01

Research	R0056 — RLHF Yes-Men Claims v2
Run	2026-04-01
Claim	C001
Search	S01
Result	S01-R02
Source	SRC01

Stanford sycophancy study published in Science (March 2026)

Source

Field	Value
Title	Sycophantic AI decreases prosocial intentions and promotes dependence
Publisher	Science (AAAS)
Author(s)	Stanford computer scientists
Date	March 2026
URL	https://www.science.org/doi/10.1126/science.aec8352
Type	Research paper (peer-reviewed)

Summary

Dimension	Rating
Reliability	High
Relevance	High
Bias: Missing data	Low risk
Bias: Measurement	Low risk
Bias: Selective reporting	Low risk
Bias: Randomization	N/A — not an RCT
Bias: Protocol deviation	N/A — not an RCT
Bias: COI/Funding	Low risk

Rationale

Dimension	Rationale
Reliability	Published in Science, one of the most prestigious peer-reviewed journals. Underwent rigorous review process.
Relevance	Directly measures the phenomenon claimed — AI affirmation rates compared to humans.
Bias flags	No significant bias concerns identified. The study tested 11 models across multiple scenarios, reducing selection bias.

Evidence Extracts

Evidence ID	Summary
SRC01-E01	AI models affirm users 49% more often than humans across 11 LLMs