Unlike most photos in this blog, the one above was not taken by me. In fact, it is not even a photo as it was generated with an AI tool. First I used ChatGPT to make a summary of a long text so that I could use it as a prompt in Blue Willow to generate an image. The prompt was an impressive good summary of the text, and after some tweaking I got an image which was acceptable (somehow there are not many photos of bookshops and instead libraries and law firms keep coming up).
In any case, interesting what can be done these days.
Input for ChatGPT:
Make a summary of following text so that I can use it as a prompt for DALL-E to generate an image. "long text here"
Output (note some mistakes in the summary, she leaves the bookshop with the note, not JC):
A man named JC goes to a bookshop in downtown Berlin to leave a booklet of his short stories on a shelf. He is worried that the book may be thrown away, but he decides to stay and watch over it. A young woman eventually picks up the book, reads it, and takes it with her. JC leaves the bookshop with a handwritten note asking the woman to bring it back when she is finished so that others can enjoy it too. The bookshop is Hugendubel in the Schlossstrasse.
Final prompt for Blue Willow after several attempts:
Generate a black and white image with a young man and a young woman in a bookshop. The woman is in the foreground and she is standing reading a book on her hands. The man is in the background, sitting on a sofa. He is watching the woman.
Not quite what I had in mind when I wrote that short story and still quite impressive considering the (lack of) effort required to generate the image.
Speacial thanks to https://www.youtube.com/@Laura_tech for pointing me to some of the tools.
