Generative AI

Generative AI models create new content—such as text, images, audio, or code—by learning patterns from very large collections of examples.

What makes it generative?

A classifier might answer “cat” when it sees a picture. A generative model can produce a new image, paragraph, or song based on patterns it learned. It generates an output one small step at a time.

Large language models work with tokens—small pieces of text. Given a prompt, the model estimates which token is most likely to come next, repeating until it has produced a response.

The anatomy of a useful prompt

Use it with judgment

Generative tools can be inaccurate, biased, or overly confident. Verify important claims, protect private information, and treat generated material as a draft that deserves human review.

• Check important facts against trustworthy sources.
• Do not paste sensitive data into tools without permission.
• Give credit and check usage rights for generated content.

Example

Weak: 'Tell me about AI.'

Better: 'Explain machine learning to a beginner in 120 words.
Use one everyday analogy and end with three key terms.'