Core concepts
Generative AI
AI systems that create new content — text, images, audio, video, or code — rather than just classifying or predicting from existing data.
01 ——
In plain English
Generative AI refers to AI models that produce new content based on a prompt, in contrast to "discriminative" AI that classifies, scores, or predicts from inputs. The wave of products since 2022 — ChatGPT, Midjourney, Sora, Suno — are all generative AI.
Main types:
- Text — LLMs like GPT-4, Claude, Gemini
- Image — DALL-E, Midjourney, Stable Diffusion (diffusion models)
- Video — Sora, Runway, Veo
- Audio / Music — Suno, Udio, ElevenLabs
- Code — GitHub Copilot, Cursor, Cognition's Devin
Why now: Three things made generative AI explode: massive training data (the internet), the transformer architecture (2017), and cheap parallel compute (GPUs). The combination crossed a quality threshold around 2022 with ChatGPT.
02 ——
Related terms
AI
Artificial Intelligence — software that performs tasks normally requiring human intelligence, like understanding language, recognising images, or generating content.
LLM
Large Language Model — the type of AI behind tools like ChatGPT and Claude, trained to understand and generate text.
Foundation Model
A large, general-purpose AI model trained on broad data that can be adapted (via prompting or fine-tuning) to many downstream tasks.
Diffusion Model
The type of AI model behind most modern image and video generators — it learns to create content by reversing a noising process.
Transformer
The neural network architecture introduced in 2017 that powers nearly every modern LLM, image generator, and AI breakthrough.
Back to glossaryLast reviewed May 2026