what sign up
nixm nixm

← learn

AI vs machine learning vs deep learning: nested, not rivals

October 14, 2026

They are circles inside circles. Artificial intelligence is the broad goal: machines doing things that need intelligence. Machine learning is the main way we get there now: systems that learn patterns from data instead of following written rules. Deep learning is machine learning done with large, many-layered neural networks, and it is what made the last decade's progress possible. Generative AI is deep learning used to make new text, images and video. So ChatGPT is generative AI, which is deep learning, which is machine learning, which is AI.

The picture

Draw four circles, each inside the last. The outer one is artificial intelligence. Inside it, machine learning. Inside that, deep learning. Inside that, generative AI. Nothing in an inner circle is outside an outer one. That is the whole answer; the rest is what each circle means.

Artificial intelligence: the goal

AI is any system that does something we would call intelligent if a person did it: understand language, recognise a face, plan a route, play a game, make a decision. The term is about the outcome, not the method. A 1990s chess program with hand-written rules is AI. So is a model that writes poetry. The field is seventy years old; the methods have changed completely.

Machine learning: the method

Machine learning is AI where the behaviour is learned from data rather than programmed. You do not write the rule for spotting fraud; you show the system a million transactions labelled fraud or fine and it works out what fraud looks like. The spam filter, the recommendation that knows what you will watch next, the credit score, the map's traffic prediction: all machine learning, and most of it older and simpler than deep learning. The key move is that the system improves with examples, which written rules never did.

Deep learning: the method that scaled

Deep learning is machine learning using neural networks with many layers, trained on very large amounts of data. Earlier machine learning needed people to decide which features mattered (the amount, the time of day, the merchant). Deep networks find the features themselves, layer by layer: edges, then shapes, then faces; letters, then words, then meaning. Around 2012 this started beating everything else at vision and speech, and from 2017 a design called the transformer made it work for language at scale. Deep learning is why your phone understands you and why a model can read a contract.

Generative AI: deep learning that makes things

Most deep learning classifies: is this a cat, is this fraud, what did they say. Generative AI runs it the other way: make a cat, make a sentence, make a video. Same networks, trained to produce rather than sort. The generative AI article covers what it can and cannot make; the LLM article covers the text kind specifically.

Side by side

  • Question: is this email spam? Classic machine learning. Learned from labelled examples, a few hundred features, runs in a millisecond.
  • Question: what is in this photo? Deep learning. A convolutional network that learned its own features from millions of images.
  • Question: write me a reply to this email. Generative AI. A large language model predicting the next word, built on deep learning.
  • Question: what move wins this chess game? AI, by any method: the 1997 version used search and hand-written evaluation; the 2017 version learned by playing itself, which is machine learning.

Why the distinction matters to a normal person

Because the claims differ. "AI-powered" on a product can mean a 2005 rule. "Machine learning" means it learned from data, which raises the question whose data. "Deep learning" or "generative" means a large model, which raises the questions of what it was trained on and what it does with what you type. When a tool is generative, ask the second question before you paste anything personal. nixm's answer: nothing you type trains anything, and no transcript is kept.

frequently asked

What is the difference between AI and machine learning?

AI is the goal of machines doing intelligent things by any method. Machine learning is the main current method: systems that learn patterns from data instead of following written rules. All machine learning is AI; not all AI is machine learning.

What is the difference between machine learning and deep learning?

Deep learning is machine learning using large neural networks with many layers that find their own features in the data. Older machine learning needed people to choose the features and used simpler models.

Is ChatGPT AI or machine learning?

Both, and more specifically generative AI built with deep learning. It is a large neural network trained on text, which is machine learning, which is a form of AI.

Is machine learning the same as AI?

No. Machine learning is one way to build AI, the dominant one today. Rule-based systems and search-based systems are AI without machine learning.

What is generative AI compared to machine learning?

Generative AI is deep learning, itself a kind of machine learning, used to produce new text, images, video or code rather than to classify or predict a label.

Which should I learn first, AI or machine learning?

Machine learning is the practical core: learn how models are trained on data and evaluated, then deep learning and language models. AI as a field is the context around it.

try it now — no account, no transcript kept. even the signal dissolves in seven days.

seed a signal →

Enter your email and receive a link to sign in. No passwords, just a secure signal to connect.

why connect

  • 20 free credits to start
  • your signals, remembered
  • buy more credits anytime
  • image & video generation features
or
+