AI glossary · Ethics, safety and privacy
AI bias
AI bias is when an AI system produces results that are unfair to certain groups of people, usually because of skewed training data or design choices.
Also called algorithmic bias, bias, fairness in AI
In more detail
AI learns from historical data, and that data often reflects past inequalities. A model can then repeat or even amplify them, for example by favouring certain names in job applications or working less accurately for some skin tones.
Reducing bias means using more representative training data, testing results across different groups and keeping people involved in high-stakes decisions.
Example
A hiring tool trained on a company’s past hires, who were mostly men, learns to rank women’s CVs lower.
Related terms
Browse related tools
AI tools listed on BroUseAI in these categories.
