Session 2 · AI Art Crafting · Lesson 04
Feedback Loops
Workflow A
Output becomes new input. Repetition produces drift, drift reveals system bias, and structure changes meaning over time.
Concept
- A loop is the simplest machine for making time visible.
- Drift is data: it shows what the model regresses toward.
- The loop's structure — not its content — carries the meaning.
If repetition produces drift, what does your drift reveal?
Student activity
- 01Run a short feedback loop (3+ iterations) — describe with ChatGPT, regenerate with Leonardo — and record each stage in the timeline.
Tools for this loop
ChatGPT
Multimodal chat with a free tier. The workshop's default interpreter — give it an image and it will describe, read, and re-read what it sees.
Sign in to the free tier, upload your image, and ask it to describe what it sees. Copy the description back verbatim — keep its mistakes; they are the material.
External tool — it has its own privacy policy and may change or require an account.
Leonardo
Image generation with a free daily token allowance. The workshop's default generator for the image → text → image translation step.
Paste the machine's description of your source image as the prompt. Do not improve it. Generate, then bring the result back.
External tool — it has its own privacy policy and may change or require an account.