Closer #009

Welcome to Closer,

a weekly newsletter where I take one of my photos, ask AI to describe it, then see what kind of image it can generate from its own description—an ongoing experiment in the capabilities of AI, and what that means for humans. This is edition #009.

Last week, I accidentally shared the output image as the input in the email I sent. Obviously the image at the top of the email was the original input image. This issue has been resolved on the website. Apologies for any confusion.

This week, I've significantly shrunk the prompt and got a shrunken down description as a result as well. What is does for the output image? Let's find out.

The order in which we operate: human experience → photograph → AI description → synthetic image. Prompts are in the grey callout boxes, the AI description is cursive.

LLMs used

GPT-4o for image interpretation and image generation.

The input

The output

The image is a striking photograph taken from inside a small, dimly lit bathroom. The foreground features a white wall-mounted sink, a soap dispenser, a black liquid soap container, and beige wall tiles. A mirror and hand dryer are visible on the right side of the frame, contributing to the utilitarian and slightly worn look of the interior.

The true focal point, however, is the small rectangular window above the sink. Through this window, brightly lit by natural sunlight, we see a serene, contrasting outdoor scene: a woman stands at the edge of a dock, facing a deep blue body of water with barren hills or mountains in the background. She wears light-colored, loose-fitting clothes and appears to be looking out over the water, possibly lost in thought.

The photograph plays powerfully with contrast—between inside and outside, darkness and light, isolation and openness. It creates a layered narrative and invites the viewer to consider the relationship between confinement and freedom, or the mundane and the sublime.