AI-generated describes text, images, audio, video, code, or data that an artificial intelligence model produced, usually from a prompt or from patterns it learned in training. A chatbot reply, an image made from a text description, and a draft email written by an assistant are all AI-generated content. The output reflects what the model learned, not a record of something a person observed or wrote.
Because generated output can look like human work, many platforms and organizations label it, and some regulations require disclosure in certain contexts. AI-generated content can also contain errors that read as confident statements, so people usually review it before relying on it.
In data work, AI-generated data, often called synthetic data, is used to add examples for training, test systems without exposing real records, or cover rare cases. Teams check generated data against the task it will serve, because realistic-looking records are not the same as useful ones.