AI glossary · Fundamentals
Natural language understanding (NLU)
In one sentence
Natural language understanding (NLU) is the part of NLP that deals with the meaning of a message: what the person wants (their intent) and which specific details they give (entities such as dates, names or products).
Also known as: NLU · language understanding · natural language interpretation · intent recognition
What is natural language understanding?
Reading a message isn’t the same as understanding it. Natural language processing covers every technique for working with language; NLU is the part that answers the question “what are they asking me for?”. Its output isn’t text but a structure that a system can act on.
Put it to work
An AI agent on your WhatsApp, CRM included
From €49/mo + VAT
It replies to customers on WhatsApp and logs every conversation in the CRM.
See the WhatsApp agentA message like “hi, have you got a table for four on Saturday night?” becomes:
| Element | Value |
|---|---|
| Intent | Book a table |
| Party size | 4 |
| Day | Saturday |
| Time slot | Evening |
Without that translation into data, a chatbot has no idea what to do with the sentence.
Related to this term
UTIO WhatsApp
- 3 users and 500 AI conversations
- CRM and shared inbox included
- Built from your website in a minute
- 14-day free trial, no card needed
From €49/mo + VAT
See the WhatsApp agentHow it works
NLU solves three problems:
- Detect the intent: the user’s goal (book, cancel, ask a price, complain).
- Extract entities: the specific data (dates, times, quantities, product names) and normalise them, for example turning “Saturday” into an actual date.
- Resolve ambiguity using context: “change it to the other one” only makes sense if you know what was being discussed.
Until recently this was done with models trained on hundreds of example phrases per intent. If a customer used a wording nobody had anticipated, it failed. Today’s language models cope with paraphrases, typos and messages containing several requests at once, which is the norm on WhatsApp.
NLU is one building block of conversational AI: without understanding there’s no useful dialogue.
Example
A restaurant in York gets WhatsApp messages like “hiya! 4 of us sat night, any room? one is coeliac”. NLU detects the intent (book), extracts the party size, the day, the time slot and a special requirement (gluten-free). The system checks availability, suggests a time and adds the allergy note to the booking so the kitchen sees it.
NLU at UTIO
UTIO’s WhatsApp agent understands messages the way people actually write them, with no buttons or menus, and turns each conversation into something actionable: a CRM record, a pipeline stage, a calendar appointment, a task or a ticket. If you already use HubSpot, Zoho, Pipedrive or Holded, what it understands and logs is written there too, in mirror mode.
No buttons doesn’t mean no structure, though: behind every reply there’s the same intent-and-entity logic described above. There’s no intent model to train: the agent works from your instructions and your business information, and you decide how much autonomy it gets.
Frequently asked questions
What's the difference between NLU and NLP?
NLP covers everything a machine does with language, including generating or translating text. NLU is the understanding part: turning a message into something actionable, such as 'wants to book, Friday, two people'.
What is an entity in NLU?
It's a specific piece of data in the message that the system needs in order to act: a date, a time, a name, an order number, a town. The intent says what the user wants; the entities say with which details.
Do you still need to train an NLU model with today's language models?
Not in most cases. You used to train it with hundreds of example phrases per intent. Today a language model understands intents from a description and a handful of examples.
Related terms
- Natural language processing (NLP)
Natural language processing (NLP) is the branch of artificial intelligence that lets machines work with human language: read, analyse, classify, translate and produce text or speech the way people use it.
- Intent
An intent is the goal behind what a person writes or says to a chatbot or AI agent, such as booking, cancelling or asking for a price. Recognising it lets the system decide what to do with the message.
- Chatbot
A chatbot is a program that holds text conversations with people, on a website or on WhatsApp, to answer questions or guide them through a task. It either follows fixed rules or uses a language model to understand what people write.
- Conversational AI
Conversational AI is the set of technologies that let a machine hold a conversation in natural language, by text or voice: understand what is said, keep track of the context and reply in a coherent way.
- LLM
An LLM (large language model) is an AI model trained on vast amounts of text to predict the next word. That single skill lets it draft, summarise, translate, classify and hold a conversation.
Updated on