01_About this Image Gallery

In September 2023, AI4Media, in partnership with Better Images of AI and AIxDESIGN launched [CLOSED] Better Images of Responsible AI with AIxDESIGN. Submissions are now closed.

The Open Call was focused on creating new stock imagery to publish in the Better Images of AI image library. We believe there is mutual benefit to create an image garden of sorts – sketches of AI that can help seed inspiration for continued better images and conversation about AI.

For the next two months, October – December 2023, AI4Media will offer 3 micro-grants of €300 ****for images submitted to Better Images of AI Image Garden around the Topics detailed below. We invite all artists, imagemakers, researchers, and community members to submit sketches, renders, collages, or whatever visual representation your heart desires to the Images of AI – AIxDESIGN Community Gallery.

In December 2023, 3 artists will be given a €300 honorarium for their contribution. All garden images will be available for free and public use under the Creative Commons by 4.0 license. Favorites will be selected by the team (Tania Duarte, Rasa Boyce, Ploi Flynn, Nadia Piet) based on aesthetic quality, requirements for the library, and community upvotes.

Below we’ve detailed the topics we’re especially interested in to seed this image garden.

02_Topics to Visualize

Topic 1 / Algorithmic Harms

Our working definition this term is:

<aside> 🧚🏽 Algorithmic Harm is understood as the exploitation of an absence of information or of behavioral biases, that can disproportionately affect members of identifiable groups, including women and people of color. According to Joel Feinberg, someone is harmed when their interests are set back, in this case as a direct result of an algorithmic decision.

Related terms:

Algorithmic wrongs, Algorithmic discrimination, Algorithmic bias

</aside>

How this concept is visualised now:

Untitled

Christina Aminashaun

Klaus Vedfelt / Getty Images

https://www.youtube.com/watch?v=Ok5sKLXqynQ

Nuances that could serve as inspiration for Better Images:

Algorithmic Harms and Algorithmic Wrongs by Nathalie DiBerardino and Luke Stark, Data & Society
Harms, wrongs, justice are culturally contextual concepts. How can this show up in our images?

Topic 2 / Supervised vs. Unsupervised machine learning

Our working definition of this term is: