Computers can learn from data. Machine learning is a kind of statistical inference that finds patterns in data, and it needs huge amounts of training data, usually supplied by people. Large language models predict the next word in a sentence. A model can be as simple as a table of patterns counted from training data. AI systems depend on their training data, so changes during training can change how a system performs, and training data can go stale. Training generative AI takes large amounts of data, which can include personal data, and most model developers do not say exactly which data sources they used. Biased training data can leave some people less well served than others.