Ai generated image featuring two painterly female figures

Decentering Ethics with AI Art

When: Thu, 21 Sep 2023 3:30 PM – 4:30 PM AEST

Speakers: Dr. Jasmin Pfefferkorn and Emilie K. Sunde

Format: 30 minute presentation & 30 minute open discussion via Zoom

 

REGISTER NOW

 

How do artists and cultural institutions shape thought and action in the space of soft ethics?

How can we think a more decentered ethical approach that considers non- human agents with AI Art? Will AI soon need some form of ethical protection in human-machine assemblages?

By decentered ethics we mean that we are interested in how ethics can be embedded in relationships and processes that account for AI’s materiality, ecological impact, and emphasize human entanglement in the technical. To arrive at this formulation we have made two interdisciplinary moves:

First, we move beyond the regulatory frameworks normatively associated with the field of digital ethics. In this, we take inspiration from Luciano Floridi’s (2018) conceptualisation of ‘soft ethics’: a post-compliance ethics that addresses what should be done beyond what is specified within government or policy frameworks. The speed of technological change is incongruous with the time required to make legal and governance decisions within a democratic system. Therefore business, institutions and communities will always be required to look beyond regulation for ethical guidance in their application of AI.

Our second move is informed by the field of ecological ethics, which pushes beyond the human-focused stance that commonly characterises scholarly studies of ethics. We name this ‘decentered ethics’, an approach that dislodges humans as the single locus of protection against ethical harms in human-machine assemblages. De La Bellacasa argues that under neoliberalism ethics is often framed as personal moral responsibility rather than a site of collective, political transformation: for example, people taking shorter showers substitutes for strategic moves to close coal power stations. Yet isolating single moral actors fails to recognise that ethics take place in unpredictable and emergent processes shaped by sociotechnical forces, practices and agencies (De La Bellacasa 2017, 138). While ethical fears around automation in sectors such as agriculture lament a reduction in human jobs, such systems only thrive where economic pressures drive a need for cheap food. Similarly, focusing on the human in the supply chain misses the cumulative effects on non-human beings, for example more ‘efficient’ farming decisions can lead to a lack of biodiversity and a loss of ecosystem health.

About the Speakers: Dr. Jasmin Pfefferkorn and Emilie K. Sunde

Dr. Jasmin Pfefferkorn and Emilie K. Sunde are researchers with the Centre for Artificial Intelligence and Digital Ethics’ (CAIDE) Art, AI and Digital Ethics research collective. Both Jasmin and Emilie work in the School of Culture and Communication at The University of Melbourne, on the ARC Project ‘Digital Photography; mediation, memory, and visual communication’. They are also the co-founders of the research group CODED AESTHETICS, which explores the ways in which computational processes are shaping our aesthetic (sensing and sense-making) experience.

To date, there has been no single volume that explicitly explores art as a way of doing AI ethics with complex human-machine entanglements. Taking AI art as a method, rather than an autonomous generative output, this edited collection will explore the relationship between artistic practices, ethics, and AI.

Our book argues that artists and cultural institutions are a vital force in the construction of a relational, collectively held ethics of human-machine assemblages. Technological change always out-paces ethical governance, producing an uncertain zone between what machines can do, and what is upheld as ethical by diverse publics. Working quickly and often provocatively, artists trace ethical tensions as they are emerging in public consciousness, providing a vital speculative approach to AI futures.

https://jasminpfefferkorn.com

https://emilie.ai/