Three terms used interchangeably and meaning different things.
ARTIFICIAL INTELLIGENCE
The broadest term. Any system performing tasks associated with human intelligence.
Includes rule-based systems written by hand, which involve no learning at all.
MACHINE LEARNING
A subset. Systems that learn patterns from data rather than following rules written by a programmer.
Most of what is called AI today is machine learning.
DEEP LEARNING
A subset of machine learning, using layered networks loosely inspired by biological neurons.
Responsible for most recent advances: image recognition, speech, and language models.
THE RELATIONSHIP
Deep learning sits inside machine learning, which sits inside artificial intelligence.
WHY IT MATTERS
Someone selling you an AI solution may mean a rule-based system from 1995 or a modern language model.
Ask what it actually does.
IN PRACTICE
The distinction matters less than what the system can reliably do for your situation.