Creating with Artificial Intelligence:

The 2024 Chantier IA Residencies at Sporobole

In the first year of the Chantier IA, several artists had the opportunity to test and experiment with artificial intelligence in the context of artistic creation. Here’s a brief overview of what was accomplished.

Marie-Ève Levasseur explored the hybridization of insects and technology, creating images using both “text-to-image” and “image-to-image” processes. Her project also involved integrating photos of objects into the process, allowing her to redirect the results of the image generation.

Dayna McLeod created avatars based on her own image. These avatars required the use of synthetic voices, the creation of conversational agents, and an application designed to modify appearance (usually used for trying out makeup or hairstyle changes).

To test the limits of AI, Simon Laroche pushed image generation (image-to-image) to its limits, repeating the process until the results stagnated and ceased to evolve. He also experimented with generating images representing words and writing, with the goal of developing a typeface specifically designed by AI.

Sabrina Ratté and Roger Tellier-Craig worked as a duo on an installation project. Sabrina generated images from text prompts as well as from personal archive photos, which were then used to generate video. Roger, on the other hand, worked on creating music and sound design from text-based instructions.

Félix Bernier, aiming to represent communication in the plant world, transformed sound recordings into visualizations. His project involves converting the real-time sound data of a location into a series of symbols invented by AI. The project offers both a three-dimensional representation of the location in a pointillist style and a translation of sound into graphic representations in the form of lines.

Two residencies are still ongoing this fall: those of Frédérick Maheux and Adam Basanta.

Frédérick Maheux is working on generating images from both text prompts and images. The numerous images produced will be used to create an interactive, installative artwork. His work highlights the over-the-top and uncanny aesthetics that AI often produces.

Meanwhile, Adam Basanta’s residency is in partnership with the Faculty of Law at the University of Sherbrooke. He is focused on developing a project that offers a critical perspective on AI and its commercial development.

Despite the diversity of projects, AI integration rarely replaced the artist’s role in decision-making and aesthetic approach. While production sometimes became faster, this often came with the challenge of selection, as artists had to act as curators, sorting through a massive volume of images. AI tools for creating music proved to be far less advanced than those for visual creation. Although most projects relied on generative AI, artists encountered issues related to biases and tendencies in these tools. In the end, this is only the beginning of a broader reflection on the evolving status of the artwork and the artist in the age of artificial intelligence.