AI Terminology
This glossary explains common AI terms in plain language. These words come up often when using tools like ChatGPT, Gemini, or Copilot.
A
AI (Artificial Intelligence): Technology that can do things that normally need human thinking—like writing, planning, or answering questions.
Automation: Letting technology handle a task for you. AI can help automate writing emails, scheduling meetings, or creating reports.
B
Bias: Sometimes AI gives unfair or unbalanced results because of patterns in its training data. It might favor one idea or group without meaning to.
C
Chatbot: An AI tool you can talk to through typing. ChatGPT is one example.
Context Window: The limit to how much the AI can “remember” in one session. If the conversation gets too long, it might forget earlier parts.
D
Data Privacy: Keeping your personal or work info safe when using AI. Don’t share passwords, private data, or anything confidential with a chatbot.
F
Fine-tuning: Giving the AI extra training to help it do better in a certain topic or task.
G
Generative AI: AI that creates things—like writing, code, images, or music. ChatGPT is generative because it writes new content.
H
Hallucination: When the AI gives an answer that sounds real but isn’t true. It doesn’t mean to lie—it just gets things wrong sometimes.
I
Inference: The process the AI uses to figure out a good response based on your prompt.
L
LLM (Large Language Model): A type of AI that’s trained to understand and generate language. These models are what tools like ChatGPT are built on.
M
Model: The part of the AI that does the thinking. It reads your prompt and figures out how to respond.
N
Natural Language Processing (NLP): How AI reads, understands, and writes human language.
P
Personalization: Adjusting the AI’s response by giving it details about who you are or what you need. Example: “Write a thank-you note from a customer support rep.”
Prompt: A message or instruction you give to the AI. It can be a question, a request, or a short description of what you need.
Q
Query: Another word for your question or search request. A prompt and a query are often the same thing.
T
Token: A piece of text the AI reads. Words and punctuation are broken into tokens. The more tokens, the longer or more detailed the response.
Training Data: All the text the AI learned from—like books, articles, and websites. This helps it understand how language works.
U
Use Case: A real example of how AI is used. Example: Using AI to summarize long emails is a common use case.
W
Workflow: A set of steps you take to get something done at work. AI can help make some steps faster or easier.