The quality, size and variety of training data shape what a model can do and what biases it has.
Example
A video model trained on lots of nature footage is often good at water and animals.
Why it matters to you
It explains why models are great at some subjects and weak at others.
Related terms
Machine LearningThe way most AI learns: by finding patterns in examples instead of following hand-written rules. BiasWhen an AI system gives unfair or one-sided results because of patterns in the data it learned from. Fine-tuningTraining an existing AI model a bit more on specific examples so it gets better at a particular task or style.