AI Hiring Bias Glossary
Comprehensive definitions of key terms in AI hiring bias, compliance, and algorithmic fairness. Your reference guide for navigating the regulatory landscape.
A
Adverse Impact
A substantially different rate of selection in hiring that disadvantages members of a protected group.
Bias & FairnessAI Interviewing
The use of artificial intelligence to conduct, evaluate, or assist in job interviews, including video analysis, speech analysis, and automated question generation.
TechnologyAlgorithmic Fairness
The principle that algorithms should produce equitable outcomes across different demographic groups.
Bias & FairnessApplicant Tracking System (ATS)
Software used by employers to manage job applications, track candidates through the hiring pipeline, and organize recruitment workflows.
TechnologyAudit Trail
A chronological record of all activities, decisions, and changes related to an AI hiring tool, used to demonstrate compliance and enable investigation.
Legal & ComplianceAutomated Employment Decision Tool (AEDT)
Any computational process that uses machine learning, statistical modeling, or data analytics to substantially assist or replace human decision-making in employment.
Legal & ComplianceB
C
Candidate Scoring
The process of assigning numerical scores to job candidates based on AI evaluation of their qualifications, assessments, or interview performance.
TechnologyChi-Squared Test
A statistical test used to determine whether observed differences in selection rates are statistically significant.
Statistical MethodsConfidence Interval
A range of values within which a true population parameter is expected to fall, given a specified level of certainty.
Statistical MethodsD
Demographic Parity
A fairness criterion requiring that selection rates be equal across all demographic groups, regardless of qualifications.
Bias & FairnessDisparate Impact
Employment practices that are facially neutral but have a disproportionately negative effect on a protected group.
Legal & ComplianceF
Fairness Metrics
Quantitative measures used to evaluate whether an AI system produces equitable outcomes across different demographic groups.
Bias & FairnessFisher's Exact Test
A statistical test for significance used when sample sizes are too small for the chi-squared test.
Statistical MethodsFour-Fifths Rule (80% Rule)
A guideline stating that a selection rate for any group should be at least 80% of the highest group's rate.
Bias & FairnessI
M
N
P
Predictive Parity
A fairness criterion requiring that an AI model's predictions are equally accurate across all demographic groups.
Bias & FairnessProtected Class
A group of people sharing a characteristic protected by anti-discrimination law, such as race, sex, age, or disability.
Legal & ComplianceProxy Variable
A seemingly neutral data point that correlates with a protected characteristic and can perpetuate discrimination.
Bias & FairnessR
S
T
Ready to Audit Your AI Hiring Tools?
Get your compliance report in minutes. No consulting engagement needed.
Start Your Free Audit