AI Ethics Principles in Practice: Perspectives of Designers and Developers

TL;DR

The paper addresses the gap between high-level AI ethics principles and practical techniques for designing responsible AI systems. It employs semi-structured interviews with CSIRO researchers to examine alignment with the Australian DISER 2020 principles, revealing concrete practices, tensions, and suggestions across eight principles. Key contributions include empirical insights, a discussion of implementation challenges, and organizational and design-oriented recommendations such as data planners, datasheets, and model cards to strengthen accountability and transparency. The findings underscore that mere principles are insufficient without supportive governance, risk assessment, and developer-focused design guidance to enable responsible AI in real-world projects. The work thus offers actionable guidance for researchers, developers, and organizations aiming to operationalize AI ethics in practice.

Abstract

As consensus across the various published AI ethics principles is approached, a gap remains between high-level principles and practical techniques that can be readily adopted to design and develop responsible AI systems. We examine the practices and experiences of researchers and engineers from Australia's national scientific research agency (CSIRO), who are involved in designing and developing AI systems for many application areas. Semi-structured interviews were used to examine how the practices of the participants relate to and align with a set of high-level AI ethics principles proposed by the Australian Government. The principles comprise: (1) privacy protection and security, (2) reliability and safety, (3) transparency and explainability, (4) fairness, (5) contestability, (6) accountability, (7) human-centred values, (8) human, social and environmental wellbeing. Discussions on the gained insights from the interviews include various tensions and trade-offs between the principles, and provide suggestions for implementing each high-level principle. We also present suggestions aiming to enhance associated support mechanisms.