The NLP task of finding named things in text and classifying them as people, organisations, locations, dates, and products; the traditional pipeline sense of an entity.
Named Entity Recognition (NER) is the natural-language-processing task of finding named things in text and sorting them into categories: people, organisations, locations, dates, products, and the like. A classifier reads a sentence and labels each span, turning free text such as "DEJAN, founded by Dan Petrovic in Brisbane" into tagged entities (organisation, person, location).
It is a foundational step in how machines turn prose into structure. NER feeds knowledge graphs, powers brand-mention detection and monitoring, and underlies the older, pipeline sense of an entity: a specific real-world thing extracted and linked. This is the traditional, technical meaning, narrower than the working definition we use for AI SEO.
For AI visibility, NER is the mechanism that decides whether your brand is even seen as an entity in a body of text. When a system extracts the entities from a page or an answer, a brand that NER recognises and resolves correctly can be counted, linked, and associated; one it misses or mislabels drops out of the graph. We rely on the same extraction when we map a brand's brand association network through model probing.
