---
id: "claim-explanations-increase-override"
type: "claim"
source_timestamps: ["\\\"§ Sometimes", "people don’t want to know.\\\""]
tags: ["human-in-the-loop", "decision-making"]
related: ["concept-algorithmic-override", "action-encourage-second-guessing"]
confidence: "high"
testable: true
speakers: ["Alex Chan"]
sources: ["adoption"]
sourceVaultSlug: "hbr-seg-adoption"
originDay: 9
articleStem: "hbr-edu-37-employees-not-questioning-ai"
sourceUrl: "https://hbr.org/2026/06/employees-arent-questioning-ai-advice-enough"
sourceTitle: "Employees Aren’t Questioning AI Advice Enough"
---
# Viewing AI explanations increases the rate of algorithmic override

**Confidence:** high · **Testable:** yes · **Attributed to:** [[entity-alex-chan|Alex Chan]]

When users overcome their willful ignorance and actually view the reasoning behind an AI's prediction, they are more likely to challenge it. **Participants who viewed explanations were about six percentage points more likely to override the AI's recommendation and approve both loans**, demonstrating that transparency — when engaged with — effectively stimulates critical human judgment.

This is the mechanism behind [[concept-algorithmic-override]] and the justification for [[action-encourage-second-guessing]]. It is the constructive mirror image of [[concept-willful-ignorance-in-ai]]: the harm is not explanations themselves but their avoidance.

**Enrichment note:** The causal link is explicitly documented — when explanations revealed the AI penalized non-White or female borrowers, override of the profit-maximizing recommendation increased. Chan also emphasizes that people undervalue explanations even when those explanations complement private information and improve accuracy, implying that engaged explanations do change decisions. **The "about six percentage points" statistic is not directly verifiable from summaries alone and should be treated as provisional.**