ChatGPT is a chat assistant from OpenAI built on its GPT family of models. Like other chat models, it predicts text and runs inference to answer, so the same token-efficiency ideas apply here.

By default it tends toward a helpful, fairly thorough style and will often produce structured, medium-length answers. That default verbosity is convenient but can spend output tokens you did not ask for.

Across this track, the focus is only what is specific to ChatGPT. The general prompting skills live in the other courses and are not repeated here.