Diffusion Model
The kind of AI that makes images by starting with static noise and gradually cleaning it into a picture.
In everyday terms
Midjourney, DALL·E and Stable Diffusion work this way: each step removes a little noise, guided by your text prompt.
For professionals
Learns to reverse a gradual noising process; sampling iteratively denoises from Gaussian noise, conditioned on text embeddings.
Think of it like…
A sculptor seeing a statue in a block of marble and chipping away, step by step.
You've already seen it
AI-generated images and videos; "steps" setting in image tools.
Myth vs reality
Myth: Image AIs stitch together pieces of existing pictures.
Reality: They generate pixels from noise, guided by learned patterns.
Quick check
A diffusion model starts generating an image from…
- A blank white canvas
- Random noise
- A stock photo
- A sketch you upload
Show answer
Random noise: It denoises random noise into an image.