# Natural Language Understanding (NLU)

> The part of language AI that finds what text means. Then a program can act on it.

**Field:** Artificial Intelligence  
**Category:** NLP & Language  
**Also known as:** NLU, Language Understanding

## Think of it like

It is like a waiter who hears your order. He knows which dish you want.

## Definition

Natural language understanding, or NLU, is the part of NLP that works out what a piece of text means. It turns words into something a program can act on. That might be a goal, a label or a list of facts. A request to book a table for two at 8 becomes a booking, with a time and a count of people. Finding the mood of a text or the names in it are NLU tasks too. It is the reading half of language AI.

## Example

Lucy types 'cancel my 3 pm meeting' into her calendar app. NLU finds the goal, cancel, and the thing to cancel, the 3 pm meeting.

## Why it matters

A program can only act on words once it knows what they mean. NLU is the step that gets it there.

## Common mistake

Thinking NLU means the computer understands the way a person does. It finds patterns that point to the meaning, and it can be fooled.

## Three principles

- **Meaning out.** Words become goals and facts.
- **The reading half.** Writing is the other half, NLG.
- **Many tasks.** Sentiment, entities, intent and more.

## Related terms

- [Natural Language Processing (NLP)](https://learn.benibix.com/term/ai/natural-language-processing)
- [Natural Language Generation (NLG)](https://learn.benibix.com/term/ai/natural-language-generation)
- [Intent Detection](https://learn.benibix.com/term/ai/intent-detection)
- [Sentiment Analysis](https://learn.benibix.com/term/ai/sentiment-analysis)
- [Named Entity Recognition (NER)](https://learn.benibix.com/term/ai/named-entity-recognition)
- [Information Extraction](https://learn.benibix.com/term/ai/information-extraction)

## Told apart from

- [NLU vs NLG](https://learn.benibix.com/compare/nlu-vs-nlg): People think NLU and NLG are two names for one thing, since both deal with plain language. They point in opposite ways. NLU reads words and turns them into meaning a program can use, like a goal or a date. NLG starts from data or a goal and writes words for a person to read. A voice assistant does NLU to hear you and NLG to answer.

![A loop in two halves. On the left, the sentence 'book a table for two at 8' goes into NLU and comes out as a form with a goal, a time and a count. On the right, the form goes into NLG and comes out as the sentence 'Your table for two is booked for 8 pm.'](https://learn.benibix.com/diagrams/ai5-nlu-nlg.svg)

---

Source: https://learn.benibix.com/term/ai/natural-language-understanding
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: 2026-11-18
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
