# How AI makes pictures · AI in Bed

Beginner level · 6 min · https://www.aiinbed.com/learn/beginner/how-ai-makes-pictures

*Type a sentence, get an image. A gentle, no-maths look at how image generators turn noise into pictures, and where they fall short.*

You type "a fox reading a book under a lamp, watercolour style" and a few seconds later there it is. It can feel like magic. The idea behind it is surprisingly gentle: start with static and tidy it up, one small step at a time.

### Learning from captioned pictures

Image generators are trained on a very large number of images, each paired with a short description. From these pairs the model learns how words connect to shapes, colours, textures and styles.

It doesn't store the pictures like a photo album. It learns general patterns, such as what "fox" or "watercolour" tends to look like.

### Starting from noise

Most modern image generators are **diffusion models**. During training, the model sees pictures with random speckles of noise added, a little at a time, until they are pure static. Its job is to learn to undo that: to guess, at each step, what a slightly cleaner picture would look like.

To make a new image, it begins with pure random noise and cleans it up step by step, guided by your words, until a picture appears.

*Interactive explainer: Watch a picture emerge from noise as each clean-up step is applied.*

> It's a bit like a sculptor who sees a figure inside a block of marble and chips away, little by little. Your prompt tells the sculptor what to look for.

**Quick check:** How does a diffusion model make a new image?

A. It finds a matching photo online and copies it
B. It starts from random noise and removes it step by step, guided by the prompt
C. It draws lines one at a time like a pen plotter
D. It glues together pieces of stored photos

Answer: B. It begins with static and repeatedly predicts a slightly cleaner version, steered by your text, until a picture forms.

### Where they still stumble

Image generators can struggle with exact counts, readable text, hands and complicated scenes with many objects in specific places. They can also repeat biases from their training images, such as who they picture as a doctor or a nurse.

### The copyright question

Many training images were collected from the internet, often without asking the artists. Whether that is allowed is being argued in courts and parliaments in several countries, and the answers are not settled yet.

There are also open questions about who owns an AI-made picture. If you use these tools for work, check the tool's terms and your local rules.

**Quick check:** Which of these is a common weakness of AI image generators?

A. They can only make black and white images
B. They need a camera to work
C. They often get exact text, counts or hands wrong
D. They can't follow any style

Answer: C. Precise details like spelling, numbers of objects and fingers are still tricky, though newer models are improving.

**Key takeaways**

- Image generators learn how words relate to pictures from huge sets of captioned images.
- Diffusion models start from noise and clean it up step by step, guided by your prompt.
- They still slip on fine details and raise unsettled questions about bias and copyright.

**Go deeper (free sources)**

- [Denoising Diffusion Probabilistic Models](https://arxiv.org/abs/2006.11239) by Ho, Jain & Abbeel, 2020
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://arxiv.org/abs/2112.10752) by Rombach et al., 2022
- [Diffusion model](https://en.wikipedia.org/wiki/Diffusion_model) by Wikipedia