I have continued working with the Flusser ChatGPT and folding ‘letters’ into the planned zine (previously discussed here). However, I was not happy about how it was manifesting. But, as so often with these things, I woke up early two days ago and knew what I had to try out. Today is the first opportunity I’ve had to print some of these latest experiments. You’ll note the ‘readability’ has become less of a concern:

Here are a few:


Before playing in this way, things looked a little different:

Saying that some of the discussions I have been having may be worthy of including one way or another (although they could be inundated with ‘hallucinations’). I experimented with synthetic voice which has come a long way since working on the astronaut project. Then I used a company called Natural Voices and ironically called the webpage on which they appeared as much. A brief experiment using a voice from RunwayML sounds far closer to a ‘real person’ – and it did make me feel very sorry for voice-over actors.

Experiment for a project which includes a synthetic voice reading out an AI response to questions I have been asking it. The AI is a ChatGPT programmed to respond as closely to Vilém Flusser as possible based on text from his book Into the Universe of the Technical Image (1985). Obviously, for the zine, this audio will be of no use, however, I will come back to it if/when Maria and I take the project further.

Although the images that include text above seem more appropriate for this project than more readable renderings, I have continued to discuss the work with the AI chatbot.

Here is its latest response:

The text I provided was from Karen Barad’s Meeting the Universe Half Way (2007, p 59.) in which Butler was cited.

Discover more from Sketchbook

Subscribe now to keep reading and get access to the full archive.

Continue reading