Skip to main content

​
What is Generative AI?

​
The Big Picture

Generative AI is artificial intelligence that creates new content—text, images, video, code, music, and more. Unlike traditional AI that analyzes or classifies existing data, generative AI produces original outputs based on patterns it learned from massive datasets. Think of it this way: Traditional AI can tell you if an email is spam. Generative AI can write the email for you.

​
How It’s Different from Traditional AI

Traditional AI (Analytical):
  • Recognizes patterns in data
  • Makes predictions or classifications
  • Examples: Spam filters, recommendation systems, fraud detection
Generative AI (Creative):
  • Creates new content from scratch
  • Generates text, images, code, audio, video
  • Examples: ChatGPT, Midjourney, GitHub Copilot

​
Why It Matters Now (2025)

Three key breakthroughs made generative AI practical for everyday use:
  1. Transformer Architecture (2017) - A new way for AI to understand context and relationships
  2. Massive Training Data - Models trained on huge portions of the internet
  3. Accessible Interfaces - Easy-to-use tools like ChatGPT made AI available to everyone
The result? AI that can understand nuanced requests and generate human-quality content in seconds.

​
The Foundation: Training on Massive Datasets

Generative AI models learn by studying enormous amounts of data:
  • Text models (like ChatGPT): Trained on books, websites, articles, code
  • Image models (like Midjourney): Trained on millions of images with descriptions
  • Code models (like Copilot): Trained on billions of lines of public code
They learn patterns, relationships, and structures—then use that knowledge to create new content.
Important: These models don’t copy or memorize content. They learn patterns and generate new outputs based on those patterns.

​
What Can Generative AI Do?

Text

Write, summarize, translate, answer questions, chat

Images

Create designs, illustrations, photos, art from descriptions

Video

Generate and edit video content, animations

Code

Write, debug, and explain code in any language

Audio

Generate voices, music, sound effects

Multimodal

Combine text, images, and more in one interaction

​
Key Concepts to Understand

Prompts: Instructions you give to the AI (we’ll cover this extensively in AI 102) Tokens: How AI processes text (roughly 4 characters = 1 token) Context Window: How much information the AI can “remember” at once Hallucinations: When AI generates plausible but incorrect information Training Cutoff: The date when the model’s training data ends (it doesn’t know events after this)

​
Curated Resources

Understanding AI Course

DataCamp’s free introduction to AI fundamentals

Microsoft's AI for Beginners

Free 12-week curriculum on GitHub

Google AI Essentials

Google’s guide to understanding AI

What is Generative AI?

Google Cloud’s 22-minute explainer video

​
Next Steps

Now that you understand what generative AI is, let’s dive into the technology that powers text-based AI:

Large Language Models (LLMs)

Learn about the AI models behind ChatGPT, Claude, and Gemini