What are embeddings?

Embeddings are the reason AI document search feels different from the search boxes you're used to. You ask about something in your own words, and it finds the right document even though that document uses entirely different terminology.

Meaning as coordinates

An embedding model reads a piece of text and outputs a long list of numbers — often several hundred. Those numbers are a position: a point placed in a space where distance corresponds to similarity of meaning.

"The invoice is overdue" and "payment hasn't come through" share almost no words, but land close together. "The invoice is overdue" and "the weather is fine" land far apart.

Nobody designed those coordinates by hand. They emerge from training on enormous amounts of text, where the model learns which expressions get used in similar ways.

What this makes possible

That last one is the important one commercially. It's how you get an AI assistant that answers from your 400-page operations manual without sending all 400 pages.

An Australian small-business example

A Perth manufacturer has fifteen years of job files. A new estimator wants to know whether they've quoted anything like a current enquiry before.

Keyword search fails — the old jobs use different product names, different abbreviations, and terminology that changed twice. Nothing matches what she types.

With the files embedded, a description of the current job surfaces four genuinely comparable past jobs, none of which share her wording. The knowledge was always there; it just wasn't reachable by anyone who didn't already know what it was called.

Where it goes wrong

All of which is a reason to keep a person reading the retrieved source rather than trusting a confident summary of it.

Why it's worth knowing the word

You'll never configure embeddings directly. But when someone proposes an AI system that "searches all our documents", knowing what's underneath tells you which questions to ask: how are documents split, how do you handle superseded versions, and does the answer show its sources?

Those three questions separate a document assistant that's genuinely useful from one that confidently quotes a policy you retired in 2021.

Frequently asked questions

Do I need to understand embeddings to use AI?
No. It's under-the-bonnet machinery. It's worth knowing the concept because it explains why AI document search finds things keyword search misses — and why it occasionally returns something that feels related but isn't what you wanted.
How is this different from a normal search?
Keyword search matches characters — 'invoice' finds documents containing 'invoice'. Embedding search matches meaning, so it also finds 'bill', 'statement' and 'amount payable'. It's more useful most of the time, and less literal when you wanted literal.
Is this the same as RAG?
Embeddings are a component of RAG rather than the whole thing. RAG is the pattern of retrieving relevant material and giving it to the model; embeddings are usually how the relevant material gets found.
Where are the embeddings stored?
In a vector database, which is built to compare these number lists quickly. For most businesses this is entirely hidden inside whichever product you're using — you never touch it directly.
Does making embeddings send our documents somewhere?
Usually yes — the text is sent to a provider to be converted. So the same data questions apply as any other AI processing: which tier, which provider, and whether the material should be leaving your systems at all.

Put this to work

Ad On Group runs AI training and enablement for Australian teams through Ad On AI — a three-month, self-paced program that takes non-technical staff from their first prompts to working AI agents.

Talk to us →

Keep reading

← All resources