AI, machine learning and deep learning are nested ideas. Artificial intelligence is the broad goal of making machines do intelligent tasks. Machine learning is the main way to achieve it: models that learn from data. Deep learning is a powerful kind of machine learning that uses many-layered neural networks.
What is the difference between AI, ML and deep learning?
| Artificial intelligence | Machine learning | Deep learning | |
|---|---|---|---|
| What it is | Any technique that makes computers act intelligently | Systems that learn patterns from data instead of hand-written rules | Machine learning with many-layered neural networks |
| Scope | Broadest | A subset of AI | A subset of machine learning |
| Typical data | Any | Tables of features (numbers, categories) | Images, audio, video and text |
| Data needed | Varies | Small to medium | Usually large |
| Examples | Chess engines, route planners, assistants | Loan risk scores, spam filters, price prediction | Face unlock, speech recognition, ChatGPT and Claude |
What is artificial intelligence?
AI is the goal, not a single technique. Early AI used hand-written rules ("if the patient has symptom A and B, suggest test C"). Rules work for narrow problems but break when the world is messy, which is why most modern AI is built with machine learning.
What is machine learning?
Instead of writing rules, you give a model many examples and let it learn the pattern. Show it thousands of past loan applications with outcomes, and it learns to estimate risk for new ones. Classical algorithms such as linear and logistic regression, decision trees, random forests and gradient boosting are the workhorses of business data.
What is deep learning?
Deep learning uses neural networks with many layers. Each layer learns slightly more abstract features: edges, then shapes, then objects in an image; or letters, then words, then meaning in text. It needs more data and computing power, but it dominates tasks involving images, speech and language. Transformers, the architecture behind today's large language models, are a deep learning design. We explain them in how LLMs like ChatGPT and Claude work.
Which should I learn first?
- Programming and maths foundations: Python, basic statistics and linear algebra.
- Classical machine learning: it teaches training, validation, overfitting and evaluation clearly.
- Deep learning: neural networks, backpropagation, CNNs and RNNs.
- Transformers and LLMs: attention, tokenisation, fine-tuning and retrieval-augmented generation.
This is exactly the order in Program Zero: Phase 6 (classical ML), Phase 7 (deep learning) and Phase 8 (NLP, transformers and building LLMs).
The bottom line
Think of three circles, one inside the other: AI contains machine learning, which contains deep learning. Learn them from the outside in, and each step will make sense.
