Embedding
A way of turning text or images into numbers that capture meaning, so AI can measure how similar two things are.
When you feed an AI your content, it converts each piece into an embedding: a long list of numbers that stands in for what that piece actually means. Two pieces with similar meaning end up with similar numbers, even when they use completely different words. That is how an AI knows that "car" and "automobile" belong together, or that a customer asking about "a refund" and one asking to "return my order" want the same thing.
This is the quiet engine behind "smart" search and grouping. Instead of matching exact words, the AI compares meaning, so it can pull the right answer from your help docs or bunch similar customer messages together. You never see embeddings directly, but almost any AI that works over your own content is using them underneath.
Why it matters
Embeddings are how an AI can search your own content by meaning, so it surfaces the right answer or the right customer instead of only an exact keyword match.