Public participation in the governance of algorithmic discrimination: A study on the democratization of technology

公众参与算法歧视治理：项技术民主化研究

Abstract

Algorithmic discrimination and its harms have attracted great attention both domestically and internationally, leading to the rise of public participation in the governance of algorithmic discrimination. Public participation in the governance of algorithmic discrimination, to a certain extent, puts into practice Feinberg's democratization ideas of "design critique," especially the bottom-up approach. This approach not only represents a direct democratic model of technological governance but also provides a good opportunity to concretely examine and promote the practice of technological democratization. Public participation in the governance of algorithmic discrimination directly addresses the upstream aspects of algorithm development and auditing, thus inherently possessing a democratic nature. Although the democratization of algorithmic discrimination governance is still in its early stages, it holds the promise of opening a new chapter in human technological democracy. We must be vigilant against the significant harm that algorithmic abuse poses to democratic politics and other areas, and it is imperative to incorporate the governance of algorithmic discrimination into the democratic framework. We should promote a combination of top-down and bottom-up approaches, strengthen communication and trust between algorithm developers and the public, thereby constructing a public participation-based governance of algorithmic discrimination based on deliberative democracy and pushing the democratization of algorithmic discrimination governance to a deeper level.