What is AI bias?
Also called: algorithmic bias, bias in AI
AI bias is when an AI system gives unfair or skewed results for certain groups of people, usually because of patterns in its training data or how it was designed. A biased model can quietly repeat discrimination at a large scale while looking objective.
How does AI bias work?
If past data reflects unfair decisions or under-represents some groups, the model learns those patterns. Choices about what to measure and optimise can add more bias.
Why does AI bias matter?
AI increasingly influences hiring, lending and services. Spotting bias, and asking who is accountable, is part of using AI responsibly.
An example
Example: A loan model trained mostly on city applicants performs poorly for rural applicants it rarely saw in training.
Related terms
Learn it properly
AI bias is covered in Safety, Ethics and Careers, with short lessons, practice questions and instant feedback.