Outbreaks from the Grid

Outbreaks from the Grid

The Art + AI Research Group at UVM
employs machine learning to investigate new approaches to artistic image production. More broadly, the project explores emerging artistic practices with Machine Learning and AI while referencing an artistic lineage to artists such as Wassily Kandinsky, Jonn Cage and Yoko Ono. These artists employed instructions and systems in their non-digital artworks. For example, Kandinsky developed a science of aesthetics with the basic elements of point, line and plane; Cage used the oracle ‘I Ching’ like a computer to inform his compositional decisions; Ono wrote poetic scores that turn her audience into active participants when they follow a series of imaginative instructions. Inspired by these artists and curious about the potential for human + machine collaboration, our work emerges through growth-mutation-evolution lifecycles; our human-machine alliances expose a dynamic relationship between creation and destruction.

Outbreaks from the Grid is the premier project of the Art + AI Research Group at the University of Vermont. In this project we cultivate artworks that evolve over generations. The resulting images emerge from both human and machine interference and express the peculiar human experience with architectures of order, power and chaos. The public is invited to participate in the project through generative art performances that take place in virtual and gallery settings.

Sample of images we’ve made in collaboration machines on Fickr

The original dataset for Outbreaks from the Grid is Dancing Stars and you can view a sample of this collection here.

Read this article about us by Helen Hill

UVM Art and AI page on Facebook

UVM Art and AI Instagram feed


The Team:
Jenn Karson, lecturer Department of Art and Art History, University of Vermont
Kerime Toksu, Vermont Advanced Computing Core, University of Vermont
Syd Culbert. BA ’23
Anna Hulse, BA ’22
Ethan Davis, M.S. Data Science ‘21
Sarah Pell, M.S. Data Science ‘20
Fred Sanford, B.S. Mechanical Engineering ‘20
Yifeng Wei, B.A. Studio Art ’19,  M.F.A Parsons School of Design ’21

CatCoders funding from Department of Computer Science, UVM Spring 2020
Northeast Cyberteam/NSF funding, Summer 2020
This work was supported in part by the National Science Foundation (NSF) under award No. OAC-1659377.
Computations were performed on the Vermont Advanced Computing Core supported in part by National Science Foundation (NSF) award No. OAC-1827314.


Heard Over the Bell

In the United States civic bell towers remain iconic landmarks in large cities and small towns. There was a time when these were the tallest architectural structures – high in the air their artisan-made bells chimed to unify a community around a singular message to gather, heed warning, or celebrate. While these landmarks continue to shape the contemporary cityscape, many of their bells are gone or remain though no longer ring. If the towers do sound it may be through speakers and digital recordings. The messages that scaffold our lives today in the U.S. are scattered bell sounds and personal notifications from our cell phones, not a singular authoritative bell.

The bell tower architecture narrows before it points to the sky, a singular path to the heavens that assimilates and homogenizes; it suggests an individual and hierarchical journey. These towers strike a nostalgic hold while they inflame instincts to unfix and break them apart at the seams; to reveal and be transparent; to explore reconfigurations and flatten hierarchical structures; to serve social justice and community; to reinvent, recreate, ideate, and disrupt; to make room in the soundscape for the many voices that previously could not be heard over the bell.