We have seen that AI is, fundamentally, a prediction machine: it looks for patterns in data and uses those patterns to make predictions. In simple terms, the difference between predictive and generative AI is the outcome of those predictions - what they are used to produce.
Predictive AI uses patterns in existing data to estimate what is most likely. It might predict which product a person is most likely to buy, whether a customer is at risk of defaulting on loan payments, or which films someone may want to watch next. Predictive AI is often used in custom or business-specific systems, because the prediction depends heavily on the organisation’s own data and the specific decision being supported.
Generative AI uses prediction to create content. For example, it can write an email by predicting which token is most likely to come next, again and again, until it has produced a complete response. Generative AI is more often used as a general-purpose capability - writing, summarising, coding, explaining - though it can also be customised or grounded with company data.