TL;DRAI is the field of building systems that perform tasks we associate with human intelligence.
Artificial Intelligence spans any technique that lets machines do "smart" things — perceive, reason, decide, generate. It's an umbrella: machine learning is a subset, deep learning a subset of that, and today's large language models a recent branch. "AI" today usually means systems that learn patterns from data rather than follow hand-written rules.
Key points
Umbrella for machine "intelligence" tasks
ML ⊂ AI; deep learning ⊂ ML; LLMs are recent
Modern AI learns from data, not hand-coded rules
Perception, reasoning, decision, generation
Common mistakes
Equating "AI" only with chatbots
Assuming AI understands like a human
Conflating AI, ML, and deep learning as identical
Try it: Draw the nesting: AI ⊃ ML ⊃ deep learning ⊃ LLMs.