Embedding
An embedding is a list of numbers that stands for a piece of text, so a computer can measure how closely two pieces of text relate. OpenAI's guide says a small distance between two embeddings suggests the texts are closely related, and a large one suggests they aren't. It lists search, recommendations and grouping similar text among common uses.
Source: OpenAI API docs, Vector embeddings, checked October 29, 2026
What Embedding means for your business
Two kinds come up most. A word embedding stands for a single word, and Google's machine learning course uses word2vec, an older method, as its example. A document embedding stands for something longer, like a paragraph or a whole page.
Our reading of OpenAI's guide: one embedding stands for one stretch of text, so a paragraph that mixes your hours, a new service and a holiday sale gets one blended set of numbers, not three clear ones.
We wouldn't try to write for the math. We'd have each section of a page answer one question in plain words, and use both names buyers use for your service, the trade term and the everyday one.
If AI tools rarely bring up your pages, even for questions you answer, we'd rewrite them so each section says one thing plainly.
Last reviewed October 29, 2026. All terms
How AI answers it today
On October 29, 2026, Google's AI Overview mixed three meanings, the everyday word, the AI one and the math one, citing dictionaries, IBM, Google's course and YouTube. ChatGPT gave only the AI meaning, with an illustrative example, and cited nothing.
Our run through ChatGPT and Google's AI Overview, October 29, 2026. A sample, not what every buyer sees.
Related terms in crawlers and technical access
Find out where your brand stands.
Tell us what you're trying to figure out. We'll get back to you within one business day with a time to talk.