Sampling Bias
A type of selection bias where certain members of the intended population are systematically more or less likely to be included in the training dataset.
In plain language
When certain types of data are over- or under-represented in your collection. If your customer survey mostly captures tech-savvy users, AI trained on it won't understand less tech-savvy customers.
Why this matters
Sampling bias is a fundamental risk that your AI governance framework must address at the data collection stage. If training data does not represent your actual user population, the AI will not serve all users fairly and may expose your organisation to fairness and discrimination risks.
Relevance
StrategyIdentifying sampling bias requires understanding the full intended user population and comparing it against what is actually in your training data, a critical early governance step.
Related terms
Putting sampling bias into practice in your organisation?
Ready to transform your AI strategy?
Partner with Australia's AI strategy and governance specialists. From adoption roadmaps to ISO 42001 audit readiness.