ai + art header

AI & Art Competition

Join us for a two-round knockout AI image generation tournament.

If you’ve been paying attention to the world of AI, you’re likely aware of the current revolution in text-to-image generators like Dalle 2, Stable Diffusion, and Midjourney. These new tools allow users to generate stunning images -from surreal and fantastical dreamscapes to photorealistic portraits or film stills- with simple text prompts. Composing the right prompt, however, has emerged as its own kind of craft. This competition is a way to showcase the artistic elements of prompt composition (the human element) alongside the new technologies themselves. The event (like our AI & Art workshop) is also meant to encourage discussion, in our community, around the social and ethical issues that arise with increased access to systems that generate synthetic media.

Come to compete.  
Come to watch.  
Come to play with Dalle 2 and Stable Diffusion.  
Come to talk about AI.  
Come to eat free pizza.


When & Where:

October 19 | 5pm-7pm  
Ahmanson Lab | Leavey Library (Directions)


Want to compete?

If you already have experience with text-to-image generators, you're likely ready to compete. This technology is new to most of us and it's safe to say, most of us are not professionals yet! Plus, this competition is more about hanging out, having fun, and talking about AI.

If you'd like to get to know Stable Diffusion or practice for the competition, you can access an instance of it here. You can also stop by the Ahmanson Lab where we have local versions of Stable Diffusion set up.

If you'd like an introduction to text-to-image generators, you can join our AI & Art workshop taking place a week prior to this event.


The Competition:

Players will compete in two consecutive rounds, attempting to craft and alter prompts that build off of, and respond to, their opponents ideas while generating appealing, interesting, and engaging images.  

Competition Format:

  1. Judges will draw a card from a hat containing a random topic.
  2. The first player will have 3 minutes to craft a prompt around that topic and to generate a set of images (Stable Diffusion generates 4 images for every prompt).
  3. The second player will have 3 minutes to assess the images from the existing set and alter the existing prompt to improve them in some fashion. How might they make the current set of images more interesting, realistic, or engaging? How might they make them more fantastical, grotesque, or bizarre? How might they take the current topic/prompt in a different direction entirely? In altering the prompt, the second player may add as much text as they like but can only remove 1/2 of the words already in the prompt - thus working with, while still altering, their opponents ideas and style.
  4. The first player then has 3 minutes to repeat the process in step 3.
  5. There will be 4 volleys, so that each player will have 4 turns writing and altering a preceding text prompt.

After 4 volleys, judges will select the best of the 32 total images generated and declare a winner for the round. The winning player will move on to the next round.

Organization of Players and Rounds:

Round One: Player A vs. Player B (30 minutes)  
Round Two: Player C vs. Player D (30 minutes)  
Round Three: Winner of Round One vs. Winner of Round Two (30 minutes)


First prize is a $100 Amazon gift certificate; second prize is a $25 Amazon gift certificate.


RSVP to join the fun.

APPLY to compete in this event.

Terraforming Mars Collaboratory

Roll over image to view gallery.