Created by Moniker, Trending in the Multiverse explores novel possibilities of content production. Awestruck by the latest iteration of AI systems that are both fascinating and worrying, the project asks us to reconsider our concepts of authorship, ownership, and creativity.
To create these speculative scenarios on a daily basis the team checks RSS feeds of Bing, Al jazeera, TMZ and Wired. From those articles they select trending topics using scraped metadata and some simple heuristics. For instance, they keep track of the selected topics to prevent the same topic coming up during the following week.
With these topics they create a story outline. This outline, or prompt, is then fed into GPT-3, a language model that uses machine learning to produce human-like text, to generate a story from the outline. Once they have the story they let another machine learning based model, Stable Diffusion, create images based on the topic of the story.
The result are a series of daily stories complemented by images where each story is presented as an individual page, first setting the (generative) prompt and then inviting the visitor to engage via the “tell me!” button. We are not sure how many stories there are but the project has been running for over 25 weeks ≈ 175 days ≈ 175+ stories.
You can also follow the project on instagram and twitter.
Trending in the Multiverse | Project Page | Moniker
Project by Moniker (Luna Maurer, Roel Wouters, Thomas Boland)
Stories are written by Open AI’s GPT-3
Images are drawn by Stability AI’s Stable Diffusion
Typeface: True Sans by Bold Decisions
Thanks to Jonas Lund