









Pick Your Favorite Artist is a project hosted on EmProps, a platform for long form generative AI. It incorporates a second generative layer, a p5js-based image-to-image algorithm, which effectively makes it supergenerative, as described by Prof. Ira Greenberg.
The piece is paying homage to nine artists (painters): Hernan Bas, Cecily Brown, Dana Schutz, Nicole Eisenman, Peter Doig, Francis Bacon, Neo Rauch, Odilon Redon, and Justin Mortimer. By embedding their distinctive styles into prompts run through a SDXL pipeline, the subjects strive for contrast, presenting topics and ideas drawn from street and popular cultures.
With 45 different ‘features’ (parameters influencing the prompts either through words or images) spread across six variables, the potential output combinations would have amounted to a whooping 47,385 units. Given the platform’s performance at the project’s inception, expediting the iterating process, crucial for consolidation, became paramount. Consequently, I opted to port the entire pipeline to my own resources using a graphing tool called ComfyUI. This involved extensive research, data collection, and the setup of numerous custom nodes and algorithms before interfacing the project to the blockchain.
Drop’s page here.
The project followed with a release of 31 (at the moment of writing) curated outputs — PYFA SHORT.