Taking Stock at FAccT: Using Participatory Design to Co-Create a Vision for the Fairness, Accountability and Transparency Community
As a relatively new forum, the ACM Fairness, Accountability, and Transparency Conference (FAccT) has emerged as a pivotal platform for activists and scholars to critically examine the implications of artificial intelligence (AI) and machine learning (ML) technologies. This conference fosters dialogue among academics, civil society representatives, and government officials from diverse fields, all focused on understanding the broader societal impacts of both currently deployed and proposed technologies.
In an effort to enhance conference governance and engagement, a large-scale participatory design (PD) process was conducted, which combined a dynamic in-person CRAFT session, an asynchronous Polis poll, and the synthesis of a governance-focused report for the FAccT leadership. This innovative approach allowed participants to actively shape the substantive agenda of the conference, creating a more inclusive and reflexive governance model.
Key Components of the Participatory Design Process
The participatory design process involved several critical components:
- In-Person CRAFT Session: This interactive session gathered diverse stakeholders to collaboratively discuss and author seed statements that reflect their perspectives on fairness, accountability, and transparency in AI.
- Asynchronous Polis Poll: This online tool enabled participants to contribute additional statements and engage in a voting process, making patterns of agreement, disagreement, and uncertainty visible to all.
- Synthesis of Governance-Facing Report: The insights and outcomes from the CRAFT session and Polis poll were synthesized into a report aimed at guiding FAccT leadership in decision-making and future planning.
Empowering Voices in the AI Discourse
The participatory design process represented one of the first instances of applying PD principles to a venue that critically interrogates the societal impacts of AI technologies. By engaging participants in a meaningful way, the initiative fostered an environment where critical scholars felt empowered to voice their concerns and contribute to shaping the future direction of the FAccT community.
This participatory model not only enhances the legitimacy of the outcomes through collective input but also strengthens the community’s commitment to addressing the pressing issues surrounding AI and its societal implications. The approach champions transparency and inclusivity, crucial elements for ensuring that diverse voices are heard in critical discussions about technology.
Advancing Participatory Design Theory
Beyond its immediate applications within the FAccT conference, this work contributes significantly to the theory of participatory design. By providing a robust case study of a co-design paradigm that is both temporally and epistemologically scalable, it offers valuable insights for future initiatives seeking to integrate participatory methods into governance structures.
As AI technologies continue to evolve and permeate various aspects of society, the importance of forums like FAccT cannot be overstated. Such platforms play a crucial role in ensuring that the conversation around fairness, accountability, and transparency remains active and inclusive. The lessons learned from this participatory design process may serve as a model for other organizations aiming to engage stakeholders in meaningful ways, ultimately fostering a more equitable technological landscape.