From the series ‘U+200B’

32 of 200 works from the series U+200B. 

The 32 works are selected at random from the set upon page load. Refreshing the page will show a new random set of images.

The pieces belonging to the U+200B collection were produced by artificial intelligence, utilising the text-to-image machine learning technology known as Stable Diffusion, specifically version 1.4. The images were generated using a single Zero Width Space (ZWSP) character as a text prompt:

​

The Zero Width Space character code is U+200B in the Unicode UTF-16 character encoding. It has no width and is intended for invisible word separation and line break control.

The project investigates the text-to-image machine learning technology by using a reductio ad absurdum approach, questioning the role of authorship within the context of the machine learning art space.

The collection encompasses a total of 200 pieces, making references to the utilization of the Zero Width Space character as textual input.

Specifications

Nr of works: 200
Full color / 4:3 ratio
Title of an individual work from the series*: constructed upon perception according to the formula U+200B (ℕ∩[I,CC])

* To ensure digital archiving, a work from the series must be stored under a distinctive and fixed name. However, this name should not be used as the title of the work. The title must always be generated in real-time based on the given guideline. To reference the image in text, whether in print or digital format, a hash of the image can be calculated and included near the image as a reference.

The work referenced by the md5 hash 27823b03e666cc6a698f2743a59a4f01