R0044/2026-04-01/Q002/SRC04

Research	R0044 — Expanded Vocabulary Research
Run	2026-04-01
Query	Q002
Search	S02
Result	S02-R01
Source	SRC04

Nature Communications — False conflict and false confirmation errors in AI medical decision-making

Source

Field	Value
Title	False conflict and false confirmation errors are crucial components of AI accuracy in medical decision making
Publisher	Nature Communications
Author(s)	Various
Date	2024
URL	https://www.nature.com/articles/s41467-024-50952-3
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	Low risk
Bias: Protocol deviation	Low risk
Bias: COI/Funding	Low risk

Rationale

Dimension	Rationale
Reliability	Published in Nature Communications. Rigorous experimental design examining AI-assisted clinical decision-making.
Relevance	Directly addresses the mechanism by which AI can reinforce incorrect clinical judgments — false confirmation errors are the healthcare-domain equivalent of sycophancy.
Bias flags	Well-controlled study. Low risk across all domains.

Evidence Extracts

Evidence ID	Summary
SRC04-E01	False confirmation errors in AI-assisted diagnosis: AI explanations increase overreliance rather than improving accuracy