---
id: "prereq-algorithmic-bias"
type: "prereq"
source_timestamps: ["¶4", "\\\"§ Sometimes", "people don’t want to know.\\\""]
tags: ["ethics", "machine-learning"]
related: ["concept-moral-quandary-avoidance", "claim-bias-suspicion-increases-avoidance"]
reason: "Required to understand why participants felt moral discomfort and actively avoided explanations that might reveal demographic influence."
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"
---
# Algorithmic Bias

**Why you need this:** Required to understand why participants felt moral discomfort and actively avoided explanations that might reveal demographic influence (see [[concept-moral-quandary-avoidance]] and [[claim-bias-suspicion-increases-avoidance]]).

The text assumes familiarity with the concept that **AI systems can inherit and perpetuate human biases** (such as race or gender discrimination) based on their training data, and that discovering this bias in a professional setting creates legal and moral liabilities.

**Enrichment / adjacent literature:** Extensive work documents algorithmic bias in credit scoring, hiring, and criminal-justice risk assessment — how models reflect and amplify existing inequities. Chan's experiment operationalizes this with **race and gender penalization** as a realistic fairness concern, connecting his findings to the fairness-through-awareness debate: revealing bias in an explanation can spur overrides, which is precisely why financially aligned participants avoid the explanation.