Contestable AI needs Computational Argumentation
This addresses the issue of AI accountability and transparency for users and regulators, but it is a position paper, so it is incremental in proposing a conceptual framework rather than presenting new empirical results.
The paper tackles the problem of making AI systems contestable, as advocated by guidelines and regulations, by proposing that contestable AI requires dynamic explainability and decision-making processes, and argues that computational argumentation is well-suited to support this need.
AI has become pervasive in recent years, but state-of-the-art approaches predominantly neglect the need for AI systems to be contestable. Instead, contestability is advocated by AI guidelines (e.g. by the OECD) and regulation of automated decision-making (e.g. GDPR). In this position paper we explore how contestability can be achieved computationally in and for AI. We argue that contestable AI requires dynamic (human-machine and/or machine-machine) explainability and decision-making processes, whereby machines can (i) interact with humans and/or other machines to progressively explain their outputs and/or their reasoning as well as assess grounds for contestation provided by these humans and/or other machines, and (ii) revise their decision-making processes to redress any issues successfully raised during contestation. Given that much of the current AI landscape is tailored to static AIs, the need to accommodate contestability will require a radical rethinking, that, we argue, computational argumentation is ideally suited to support.