AI Glossary: 30+ Terms Explained Simply
Confused by AI jargon? This glossary explains common AI terms in plain English - from algorithms to transformers, no technical background required.
AI Glossary: 30+ Terms Explained Simply
AI conversations are full of confusing terms. Here's your cheat sheet - all the buzzwords explained in plain English.
The Basics
Artificial Intelligence (AI)
Software that can perform tasks that typically require human intelligence - like understanding language, recognizing images, or making decisions.
Machine Learning (ML)
A type of AI where computers learn from examples instead of being explicitly programmed. Show it 1000 pictures of cats, and it learns to recognize cats.
Deep Learning
A more advanced form of machine learning using "neural networks" with many layers. It's what powers most modern AI like ChatGPT.
Algorithm
A set of instructions for solving a problem. Like a recipe, but for computers.
Neural Network
AI architecture inspired by the human brain. Layers of connected "neurons" that process information.
Language AI
Large Language Model (LLM)
AI trained on massive amounts of text that can understand and generate human language. Examples: GPT-4, Claude, Gemini.
GPT (Generative Pre-trained Transformer)
OpenAI's series of language models. GPT-4 powers ChatGPT. "Generative" = creates content. "Pre-trained" = learned from lots of data first.
Natural Language Processing (NLP)
AI's ability to understand and work with human language - reading, writing, translating.
Chatbot
An AI program designed to have conversations with humans. ChatGPT and Claude are chatbots.
Prompt
The text you type to tell AI what you want. The input that starts the conversation.
Token
How AI measures text length. Roughly 1 token = 3/4 of a word. Models have "token limits" for how much they can process at once.
Context Window
How much text an AI can "remember" in a conversation. A bigger context window means it can reference more of your conversation history.
Hallucination
When AI confidently states something false as if it were true. It's not lying - it's generating plausible-sounding text that happens to be wrong.
Image AI
Text-to-Image
AI that creates images from text descriptions. "A cat wearing a space suit" → actual image.
Diffusion Model
The technique behind most modern image AI (Midjourney, DALL-E, Stable Diffusion). Starts with noise and gradually refines it into an image.
Stable Diffusion
A popular open-source image generation AI. Free to use and run on your own computer.
Upscaling
Using AI to increase image resolution while adding believable detail.
Inpainting
AI that fills in selected parts of an image - remove someone from a photo, add a new background, etc.
Video & Audio AI
Text-to-Speech (TTS)
AI that converts written text into spoken audio. Powers voice assistants and audiobooks.
Speech-to-Text (STT)
The opposite - converts spoken words into written text. Powers transcription tools.
Text-to-Video
AI that creates video clips from text descriptions. Tools like Sora, Runway, and Kling.
Voice Cloning
AI that can replicate someone's voice from samples. Useful but raises ethical concerns.
Technical Terms
Training
Teaching an AI by showing it examples. The AI adjusts its internal settings to get better at the task.
Fine-tuning
Taking a pre-trained AI and training it more on specific data to specialize it for a particular task.
Parameters
The adjustable settings inside an AI model. More parameters generally = more capable. GPT-4 has hundreds of billions.
Inference
When AI uses what it learned to make predictions or generate output. The "thinking" part when you ask ChatGPT a question.
API (Application Programming Interface)
A way for programmers to connect to AI services. How apps and websites integrate ChatGPT, DALL-E, etc.
Open Source
Software whose code is freely available. Anyone can use, modify, and distribute it. Stable Diffusion and LLaMA are open source AI models.
Proprietary
The opposite of open source - owned and controlled by a company. ChatGPT's underlying code is proprietary.
Deployment Terms
Cloud AI
AI that runs on remote servers. You send requests over the internet. ChatGPT runs in the cloud.
Local AI
AI that runs on your own computer. More private, but requires good hardware.
Edge AI
AI that runs on devices like phones or smart cameras, without needing the cloud.
SaaS (Software as a Service)
AI tools you access through a website with a subscription. Most AI tools are SaaS.
Ethical & Safety Terms
Bias
When AI reflects unfair patterns from its training data. Can lead to discriminatory outputs.
Alignment
Making sure AI behaves according to human values and intentions. A major focus of AI safety research.
Guardrails
Restrictions built into AI to prevent harmful outputs. Why ChatGPT won't teach you to make weapons.
RLHF (Reinforcement Learning from Human Feedback)
Training AI by having humans rate its outputs. How ChatGPT learned to be helpful and safe.
Industry Terms
AGI (Artificial General Intelligence)
Hypothetical AI that matches human intelligence across all tasks. We don't have this yet.
Narrow AI
AI that's good at specific tasks but can't generalize. All current AI is "narrow."
Foundation Model
Large AI models trained on broad data that can be adapted to many tasks. GPT-4 and Claude are foundation models.
Multimodal
AI that can work with multiple types of input - text, images, audio, video. GPT-4 with vision is multimodal.
---
Quick Reference Card
---
*Bookmark this page! Come back whenever you encounter a confusing AI term. We'll keep updating it.*
Share this article
Never Miss a Breakthrough AI Tool
Get the hottest AI tools, exclusive tutorials, and insider tips delivered to your inbox every Friday. Free forever.
🔒 No spam, unsubscribe anytime. We respect your inbox.