Reinforcement Learning
Training by trial, error and reward rather than by examples. It is how models are tuned to be helpful and to refuse things they should refuse.
Related terms
Conversational Search
Searching by asking a full question, often as a follow-up to a previous one, the way you would ask a person. It favours pages that answer plainly over pages stuffed with keywords.
Content Negotiation
When the same URL serves different formats depending on what the visitor asks for — HTML for a browser, Markdown for an assistant. It lets one address serve people and machines without either getting a compromise.
Context Window
How much text a model can hold in mind at once, counted in tokens. Anything beyond it is forgotten, which is why very long conversations start to lose earlier details.
Transformer
The model architecture behind almost every modern AI system, introduced in 2017. Its trick is paying attention to how every word in a passage relates to every other, which is what made fluent machine writing possible.
User-Agent
The name a browser or bot gives when it requests a page. It is how a server tells Googlebot from ChatGPT from a person, and how robots.txt rules are addressed to one and not another.
NLP (Natural Language Processing)
The field concerned with computers handling human language: understanding it, generating it, translating it. Everything from spam filters to chat assistants sits somewhere in it.