> ## Content Index
> Fetch the complete content index at: https://mitchellens.ink/llms.txt
> Use this file to discover other available public pages before exploring further.

# Closer #023
- URL: https://mitchellens.ink/closer/023/
- Published: 2025-09-10T18:00:41.000Z
- Updated: 2025-10-06T14:31:49.000Z
- Description: Whatever time I have left, I spend wrecking my brain over the conclusions of this soon-to-be 25 weeks of AI-experimentation. It's taking shape, albeit slowly, and this week is very helpful with that.
- Author: Mitchel Lensink
- Tags: Closer, Field

SPONSORED 

This weekly experiment is on an indefinite hold, until significant AI-advancements have been made.  
  
Curious about the conclusions? Read [Photography in the Age of AI (2025): Lessons From 25 Weeks of Experimentation](https://mitchellens.ink/closer/photography-in-the-age-of-ai/)

[Read now ](https://mitchellens.ink/closer/photography-in-the-age-of-ai/) 

Welcome to Closer,

A weekly experiment in AI interpretation: I share a photo, ask AI to describe it, then generate an image from that description. This is edition #023\. Prompts are in grey boxes. *Descriptions are in italics*.

Back in 2020 I did my first long(er) term photography project that *wasn't* a 365 project. It was my first attempt to discover a voice of my own. The project revolved around three 360 degrees panorama photos, that I mounted on the inside of steel frames. You could, quite literally, step *into* the picture. You can find out more on [the dedicated project page](https://mitchellens.ink/projects/the-silence-around/).

For this week's entry, I want to highlight one of the companion images that were presented along the panoramas in [a limited edition zine](https://shop.mitchellens.ink/products/the-silence-around) (also my first ever. I still have a few copies left, if you're interested.) It's an image I like looking at, because it represents a change in my journey. But also because it's such a quiet image, that rewards a longer look.

So I'm going to let AI butcher it this week, I guess.

## LLM models used

**GPT-5 Auto** for image interpretation, and image generation.

## The input

✍️

Please describe the attached image as well as you can. Don't skimp out on details and make sure you will be able to understand what's in the image when you read the description. You need to be able to form a mental picture of the image, without actually seeing it.

![The image shared with ChatGPT as input.](https://storage.ghost.io/c/8a/15/8a1564aa-a008-4e21-8398-12d04de07fcb/content/images/2025/09/009_Mitchel-Lensink_200804_Netherlands-Amersfoort-De-Stilte-Rondom-The-Silence-Around-1.jpg)

The image shared with ChatGPT as input.

## The output

*The photo shows the view into a small backyard or garden space. In the foreground, on the right-hand side, there is an old wooden gate made of vertical planks. The wood is dark and heavily weathered, with peeling green paint that reveals raw, lighter wood beneath. A piece of blue string or rope is tied loosely to part of the gate.*

*To the left of the gate, a simple metal garden gate with a wire mesh is visible, its frame also showing signs of wear and rust. This gate opens onto a short paved path that runs through a patch of grass. The grass looks slightly dry, not lush green, and the paved path has moss and uneven patches between the stones.*

*The background is dominated by a low brick building, likely a shed or small outbuilding, with a roof of old clay tiles in different shades of brown and red. Some tiles are uneven, adding to the rustic, weathered feel. Against the brick wall are some decorative items:*

- *A small triangular red-and-white birdhouse mounted between the window and the downspout.*
- *A metal or stone ornament shaped like an eagle or another bird in flight.*
- *A butterfly-shaped wall decoration, possibly made of plaster or metal.*
- *A floral or sun-shaped decoration beneath it.*

*There’s a narrow, raised garden bed at the base of the wall, edged with short cut logs placed vertically. It holds lavender or a similar bushy plant.*

*In front of this wall, two identical green plastic patio chairs are placed side by side. They are simple garden chairs, slightly faded from outdoor exposure.*

*Above, the sky is partly visible — lightly clouded but bright, suggesting a calm afternoon. Behind the roofline, the tops of leafy green trees can be seen. A single black utility wire runs across the scene, stretching from the roof toward the upper right of the frame.*

*The overall atmosphere is homely, quiet, and a little nostalgic — the type of backyard that has aged gently, with objects and decorations accumulated over time, radiating an understated sense of stillness.*

✍️

Begin with a concise checklist (3-7 bullets) of the steps you will take before generating the image. After the image is generated, validate in 1-2 lines that the aspect ratio is correct and that the description has been accurately interpreted; if validation fails, adjust or seek clarification. Generate an image with a 2:3 aspect ratio using this description: {**the description previously generated*}

![The image generated by ChatGPT based on its description of my image alone.](https://storage.ghost.io/c/8a/15/8a1564aa-a008-4e21-8398-12d04de07fcb/content/images/2025/09/ChatGPT-Image-Sep-9--2025-at-05_17_08-PM.jpeg)

The image generated by ChatGPT based on its description of my image alone.

## Impressions

You might notice we reverted back to the original prompt from [Closer #001](https://mitchellens.ink/closer/001/). I thought it was fitting to go back to our roots, because we're also kinda going back to my roots as an independent creative with the image from this week. It helps that the original prompt and the version we landed on after 22 iterations is presenting very similar descriptions.

I must say I quite enjoyed reading this week's description too. The mood of the original image is described so well. Almost thoughtfully. It zeroes in on the faded and calming esthetic of the image, describing each detail with great accuracy. For an image I know so well, this description of it has brought me a new appreciation for it.

For the AI-image generation of this image, we used the prompt we landed on last week again. Just to see how that would interplay with our original description-prompt. I think it doesn't matter much. ChatGPT 5 went through the checks, confirmed it did everything correctly, and proudly presented the results.

Technically, the AI has done what I asked it to. All the elements are there, the faded textures are present, the image is definitely similar to my original one. I just... don't like it. At all.

It *looks* like AI in a way you will understand if you have seen enough photo-realistic AI images by now (and you should, if you have been following along with this newsletter.) There's a painterly quality to it that doesn't look nice. But, who knows, we might come to appreciate this vibe as we did with analogue film and now, perhaps even more surprisingly so, early digicams and VHS video.

Those mediums all had their imperfections. Things to improve, to 'overcome,' so to speak. Which we did. Image quality in 2025 is outstanding in any modern camera. But now we feel nostalgic for that imperfect look. Wishing we could go back. I currently think it's ridiculous to expect the same to happen with AI images. But it'd be foolish to believe my opinion can't change on this as time progresses.

Please fill out the survey below if you can. It's anonymous so don't hold back. Also do let me know what you'd like to see in next week's issue. Which will be the last one following this format.

Thanks again for following along, see you next week.

Mitch

****Got a minute?** I’d love your quick thoughts — 3 multiple choice questions, that’s it.

[Share feedback ](https://tally.so/r/3102AW)