Projects
Paradise Unseen
Paradise Unseen
An immersive exhibition experience that explores John Milton's Paradise Lost as the source text in combination with photography and audio description to reach a diverse audience. ‘Paradise Unseen’ – challenging the lack of disability representation in AI-generated images and the Ocularcentrism of art practice.
The project looks at the contemporary relevance of John Milton's Paradise Lost, which he wrote after going blind. This stems from my interest in Milton's ability to adapt his creative practice, as I am exploring the same. We will curate a library of approximately 200 self-portrait photographs from the disabled community, embracing all its intersectionality.
Each photograph will be linked with quotations from Paradise Lost Book 1 using audio description techniques. These will be used to train an AI model to create an interactive text-to-image generator. This generator will serve as the basis for a pilot immersive experience, allowing contributing artists and collaborators to compare our socially engaged approach to text-to-image generation with mainstream tools.
To reach a diverse audience, the pilot will be designed to accept both voice and written input, providing visual and spoken output, but not relying solely on either.
Explore
£5,000
Aims of the Project
Combining disabled people’s self-portraits with descriptions linked to Paradise Lost offers a unique opportunity to merge photographic and audio description practice.
This approach considers new ways of image-making and expression that are less reliant on sight. It aims to challenge the limitations of AI image generators, which often lack the flexibility, individual interpretation and active involvement that are essential in traditional photography. By working with experts in training the model, critiquing and providing alternatives to the clichéd ‘AI art styles’ and the images they produce, a key objective is to explore the boundaries of AI-generated art.
Multi-disciplinary and collaborative working practice are crucial adaptations for this project. Collaborating with Disability Arts Online will help reach disabled artists, while Bonington Gallery’s interest in making exhibitions accessible to a diverse audience will enhance the project’s impact. Additionally, academic Miltonists, with their research into Paradise Lost’s contemporary relevance and its unique haptic quality, will provide valuable insights.
The project also seeks to contribute to public discussion and awareness of AI biases and norms, focusing on visible markers of diversity and disability, and their description. This will help foster a more inclusive and representative approach to AI-generated imagery.
How did they do that?
As a partially blind photographer and audio describer with a degenerative sight condition, I am developing ‘Paradise Unseen’ in partnership with Disability Arts Online, the Bonington Gallery, Nottingham Trent University, my local visually impaired community, and Miltonist academics.
Using John Milton’s Renaissance epic, Paradise Lost as the source text, we will combine photography and audio description to explore creating aesthetic outcomes in an approach that seeks to challenge the representational bias and style stereotypes often seen in AI-generated imagery.
Meet the artist
Karren Visser
Lead artist
she/her
Karren Visser is a partially-blind photographer experiencing ongoing sight loss. Her work involves socially engaged photography, focusing on participatory workshops and the representation of individuals and communities, particularly those with disabilities.
Karren has been involved in several notable projects, including ‘Seeing in Isolation’, a collaboration with Multistory and Sandwell Visually Impaired, and ‘Sight Lines’, a project funded by Arts Council England, aimed at enhancing her resilience and productivity, making photography accessible to a diverse audience.
Her recent participation in a Lenscratch feature on blind artists further illustrates her interest in new ways of image-making and expression less reliant on sight.
