359
Productivity & Workflow355
Software Development249
Automation & Workflow238
Marketing & Growth207
AI Infrastructure & MLOps197
Photography & Imaging169
Writing & Content Creation195
Design & Creative172
Data & Analytics165
Customer Support157
Education & Learning144
Sales & Outreach133
Voice & Speech128
Operations & Admin119
A new glossary explains common AI words like LLM, tokens, hallucinations, and AI agents, and why definitions can differ between companies.
In short: More AI coverage is using specialized vocabulary, and The New York Times published a practical glossary to help readers follow along.
AI is showing up in everyday products and news, and so are the words used to describe it. Terms like artificial intelligence, machine learning, and generative AI often get used interchangeably, even though they mean different things.
The glossary breaks down common building blocks behind today’s chatbots and image tools. For example, a large language model (LLM) is an AI system trained on lots of text, but its answers are based on patterns and are not guaranteed facts. It also defines tokens, which are small chunks of text a model reads (like counting words and punctuation), and explains that a model’s context window is the amount it can consider at one time (like how much you can see on one page).
It also covers newer terms that come up in safety and product design. “Hallucination” means the AI makes up something that sounds convincing but is wrong. “AI agent” describes a system that can take multiple steps and use tools to reach a goal, which is why access and permissions matter, since it could take an unintended action.
As more companies market AI features, definitions will keep shifting between researchers and vendors. It will be worth checking whether an AI tool is generating content, retrieving information from a source first (retrieval-augmented generation, which is like looking things up before answering), or taking actions as an agent.
Source: NYTimes