Towards Transparency by Design for Artificial Intelligence
- PDF / 865,809 Bytes
- 29 Pages / 439.37 x 666.142 pts Page_size
- 76 Downloads / 257 Views
Towards Transparency by Design for Artificial Intelligence Heike Felzmann1 · Eduard Fosch‑Villaronga2 · Christoph Lutz3 · Aurelia Tamò‑Larrieux4 Received: 11 January 2020 / Accepted: 29 October 2020 © The Author(s) 2020
Abstract In this article, we develop the concept of Transparency by Design that serves as practical guidance in helping promote the beneficial functions of transparency while mitigating its challenges in automated-decision making (ADM) environments. With the rise of artificial intelligence (AI) and the ability of AI systems to make automated and self-learned decisions, a call for transparency of how such systems reach decisions has echoed within academic and policy circles. The term transparency, however, relates to multiple concepts, fulfills many functions, and holds different promises that struggle to be realized in concrete applications. Indeed, the complexity of transparency for ADM shows tension between transparency as a normative ideal and its translation to practical application. To address this tension, we first conduct a review of transparency, analyzing its challenges and limitations concerning automated decision-making practices. We then look at the lessons learned from the development of Privacy by Design, as a basis for developing the Transparency by Design principles. Finally, we propose a set of nine principles to cover relevant contextual, technical, informational, and stakeholder-sensitive considerations. Transparency by Design is a model that helps organizations design transparent AI systems, by integrating these principles in a step-by-step manner and as an ex-ante value, not as an afterthought. Keywords Transparency · Artificial intelligence · Framework · Automated decisionmaking · Accountability · Design · Interdisciplinary · Ethics
Introduction The rise of machine learning and artificial intelligence (AI) has led to the creation of systems that can reach largely autonomous decisions, such as AI-based diagnostic tools for health applications (e.g., detection of diabetic retinopathy, cf. Abràmoff * Eduard Fosch‑Villaronga [email protected] Extended author information available on the last page of the article
13
Vol.:(0123456789)
H. Felzmann et al.
et al. 2018), recommender systems (e.g., YouTube recommender algorithms, cf. Bishop 2018), or predictive policing and criminal sentencing (Brayne 2017; Brayne and Christin 2020). While traditionally algorithms had to be programmed ‘by hand’ with rules to follow and weights to attach to specific data points, machine learning algorithms have changed the way patterns are extracted from datasets and how predictions are made (Van Otterlo 2013). In this article, we are interested in the delegation of a decision-making process to an algorithm, i.e., automated decision-making (AlgorithmWatch 2019). We understand automated decision-making (ADM) as a subpart of AI, an automated process with no human involvement to reach a decision (Karanasiou and Pinotsis 2017; ICO 2020). We start with the premise that automate
Data Loading...
