Natural Language Processing (NLP)
Natural Language Processing, or NLP, is the branch of AI concerned with human language: reading it, understanding it, and producing it. It covers everything from spam filters and search engines to translation apps, voice assistants, and chatbots. The goal is to let computers work with the messy, ambiguous, endlessly varied way that people actually write and speak, rather than the rigid commands computers traditionally understand.
Human language is far harder for computers than it looks. The same word can mean different things, meaning depends heavily on context, and people rely on tone, sarcasm, and shared knowledge without spelling anything out. Consider “I never said she took the money”, which shifts meaning depending on which word you stress. NLP is the field devoted to teaching machines to cope with all of this, so they can pull useful information out of text and speech and respond in kind.
NLP is not one single technique but a whole toolkit of tasks. Some are about understanding, such as working out whether a review is positive or negative, extracting names and dates from a document, or answering a question. Others are about producing language, such as translating between English and Welsh, summarising a long report, or holding a conversation. Modern voice assistants and chatbots stitch many of these tasks together.
For years, NLP made steady but unglamorous progress. The arrival of large language models transformed it almost overnight, taking systems that were merely serviceable and making them remarkably fluent. As a result, much of the AI industry we cover, and a great deal of the UK’s research strength, sits squarely within NLP. When a company says it works on understanding contracts, analysing customer messages, or building an assistant, it is doing NLP.