# Self-Supervised Learning

> The data makes its own answers. The model guesses a hidden word, then checks.

**Field:** Artificial Intelligence  
**Category:** Machine Learning  
**Also known as:** Self-Supervision

## Think of it like

Think of it like a word game. Hide one word in a book, then guess it. The book has the answer.

## Definition

Self-supervised learning makes its own labels from raw data. It hides part of an example and asks the model to guess the hidden part. The real hidden part is the answer, so no person has to label anything. Large language models learn this way, by guessing the next word in huge amounts of text. It turns plain data into training data for free.

## Example

A model reads the sentence 'The cat sat on the mat' with mat hidden. It guesses rug, checks against mat, and learns a little. It does this billions of times.

## Why it matters

It is how today's big language and image models learn. They read huge piles of data that no one labelled.

## Common mistake

Calling it unsupervised. It does learn from answers. The answers just come from the data itself, not from people.

## Three principles

- **Hide and guess.** Cover part of the data.
- **Data is the answer.** No people needed to label.
- **Scales up.** Works on huge amounts of text.

## Related terms

- [Supervised Learning](https://learn.benibix.com/term/ai/supervised-learning)
- [Unsupervised Learning](https://learn.benibix.com/term/ai/unsupervised-learning)
- [Semi-Supervised Learning](https://learn.benibix.com/term/ai/semi-supervised-learning)
- [Transfer Learning](https://learn.benibix.com/term/ai/transfer-learning)

## Told apart from

- [Self-Supervised vs Unsupervised Learning](https://learn.benibix.com/compare/self-supervised-vs-unsupervised-learning): People think self-supervised learning is the same as unsupervised learning, since no person labels anything. It is not. The second kind has no answers at all, and looks for groups. The first hides part of the data and makes that part the answer. So it trains the way supervised learning does, on labels it made itself.

## Taught in

- [Supervised and Unsupervised ML](https://learn.benibix.com/learn/ai-machine-learning/supervised-and-unsupervised-ml): Machine Learning, 6/18

![A sentence with one word hidden: the cat sat on the blank. The model guesses mat, and the hidden word from the same sentence checks the guess.](https://learn.benibix.com/diagrams/ml-self-supervised.svg)

---

Source: https://learn.benibix.com/term/ai/self-supervised-learning
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-17
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
