Animals: the forgotten stakeholders in AI ethics discourses?

Abstract

The development and application of AI-based technologies in general is accompanied by several ethical challenges, such as transparency, fairness, privacy, accountability, and sustainability. To address these challenges, a number of frameworks, guidelines and laws have been formulated in recent years. To build AI systems in an ethical way to benefit society, researchers and developers can draw on methods such as modified Value Sensitive Design approaches. Policy makers may for instance refer to the recently published UNESCO recommendations on the ethics of AI. What these frameworks have in common is that they focus exclusively on the benefits and risks of AI for humans. However, implementing AI in the context of animal science is even more complex. Animals are recognized as sentient beings who can interact and build relationships with humans. At the same time, they are in a vulnerable position as they rely on human care. Subjecting animals to technologies which have the potential to fundamentally change our notions and practices of animal handling and care may therefore be reason for new societal concerns. For a successful implementation of AI-based solutions in the context of animal science, it is important to stay ahead of these concerns and to be able to respond to them. This can only be achieved by analysis of and systematic reflection on the normative questions that may emerge when human-animal relations are mediated, enhanced or disrupted by AI. This presentation will explore approaches to develop and apply AI-based technologies in a responsible way for the benefit of society as a whole – including animals. I argue why this asks for a transdisciplinary approach: including animal perspectives into the design of AI-based solutions requires the expertise of relatively unrelated fields, such as animal science, computing sciences and ethics. In addition, to ensure that technologies are socially robust and contribute to the challenges stakeholders face in practice, it is essential to involve partners from industry, public organizations and governments.