Types & Fields of AI
All AI today is “narrow” — and it splits into fields like machine learning, vision and language.
What you will learn
- Tell narrow AI from general AI
- Name the main fields inside AI
- Match real products to the right field
Narrow AI vs general AI
Narrow AI is built to do one job well — filter spam, recognise a face, recommend a film. Every AI in the world today is narrow AI.
General AI (AGI) would think and learn across any task like a human can. It does not exist — it is still science fiction and research.
Watch out: When a headline says “AI is becoming human”, remember: real AI is narrow. A spam filter cannot drive a car, and a chess engine cannot write an email.
The main fields inside AI
| Field | What it does | You see it in |
|---|---|---|
| Machine Learning | Learns patterns from data | Recommendations, spam filters |
| Computer Vision | Understands images & video | Face unlock, photo tagging |
| Natural Language (NLP) | Understands & writes text/speech | ChatGPT, voice assistants, translation |
| Robotics | AI that senses and acts in the world | Warehouse robots, self-driving cars |
| Expert Systems | Follows human-made rules to decide | Medical checklists, tax software |
Machine Learning is the engine behind most modern AI, so most of this course builds toward it. The other fields often use machine learning inside them.
Tip: A quick test: if the system learns from examples, it is machine learning. If it follows fixed rules a person wrote, it is an expert/rule-based system. We will build one of each later.
Q. A photo app that recognises your friends’ faces mainly belongs to which field?
✍️ Practice
- Sort these into fields: Google Translate, a Roomba vacuum, Netflix suggestions, face unlock.
- Write one sentence explaining why a chess AI is “narrow”.
🏠 Homework
- Pick one AI field and find two real products that use it. Note what each product does.