pith. sign in

arxiv: 2604.16224 · v1 · submitted 2026-04-17 · 💻 cs.HC · cs.AI· cs.CY

"Taking Stock at FAccT": Using Participatory Design to Co-Create a Vision for the Fairness, Accountability and Transparency Community

Shiran Dudy , Jan Simson , Yanan Long This is my paper

Pith reviewed 2026-05-10 07:30 UTC · model grok-4.3

classification 💻 cs.HC cs.AIcs.CY
keywords participatory designconference governanceFAccTAI ethicscommunity participationPolisreflexive governanceco-design
0
0 comments X

The pith

Participatory design using statement authoring and Polis voting helps shape governance for the FAccT AI conference.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper reports on a participatory design process conducted at the ACM FAccT conference to co-create a vision for its governance. Participants took part in an in-person session and an online poll where they authored statements about the community's direction and voted to show patterns of agreement, disagreement, and uncertainty. The authors present this as one of the first applications of such methods to a venue focused on the societal impacts of AI, creating space for critical voices. A reader might care because it suggests a practical way for academic communities to include diverse perspectives in decision-making rather than relying solely on organizers.

Core claim

The paper establishes that a large-scale participatory design process, combining an in-person CRAFT session with an asynchronous Polis poll, allows participants to shape the substantive agenda for FAccT by authoring seed statements and voting on them. This fosters a niche where critical scholars can voice concerns about AI technologies, and it advances large-scale PD theory by providing a case study of a co-design paradigm that scales temporally and epistemologically.

What carries the argument

The participatory design process that integrates in-person collaboration with asynchronous online voting on community statements to inform conference governance.

If this is right

  • FAccT leadership gains a synthesized report from participant input to guide future decisions.
  • Critical scholars in AI ethics find a structured space to express concerns about societal impacts.
  • The method provides a scalable example for applying participatory design to similar academic venues over time and across different perspectives.
  • Patterns of voting make visible areas of agreement and uncertainty within the community.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Other conferences examining technology's societal effects could adopt similar processes to increase inclusivity in their planning.
  • Highlighting uncertainty in votes might encourage further dialogue on contentious topics in AI governance.
  • The approach could be extended beyond one event to maintain ongoing community input between conferences.
  • This case might inspire adaptations in non-academic settings like policy forums on AI.

Load-bearing premise

The participatory design process, including statement authoring and Polis voting patterns, produced representative and actionable input for FAccT governance without significant participation bias or synthesis artifacts.

What would settle it

If a majority of FAccT participants report that the resulting report did not reflect their views or had no impact on conference governance, that would indicate the process failed to deliver representative input.

Figures

Figures reproduced from arXiv: 2604.16224 by Jan Simson, Shiran Dudy, Yanan Long.

Figure 1
Figure 1. Figure 1: The Participatory design process: participants co-created seed statements with researchers, then engaged with the [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Participation over time as measured by the cumulative rate of new participants casting their first vote (left [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Participant voting behavior by number of votes cast (a) and overall voting distribution (b). [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Engagement of participants who contributed statements. Individual statement creators are spread along the [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Participation engagement over time. Votes received per statement (a) and duration of participation (b). [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Classification of statements as normative and descriptive. [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Distribution of all 59 statements across the 11 identified themes. [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Successful and unsuccessful examples of active deliberation by participants: (a, b) direct responses, (c) response with [PITH_FULL_IMAGE:figures/full_fig_p012_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Participation intervals of all participants, represented by dashed line segment from the time of the first vote cast [PITH_FULL_IMAGE:figures/full_fig_p027_9.png] view at source ↗
read the original abstract

As a relatively new forum, ACM FAccT has become a key space for activists and scholars to critically examine emerging AI and ML technologies. It brings together academics, civil society members, and government representatives from diverse fields to explore the broader societal impacts of both deployed and proposed technologies. We report a large-scale participatory design (PD) process for reflexive conference governance, which combined an in-person CRAFT session, an asynchronous Polis poll and the synthesis of a governance-facing report for the FAccT leadership. Participants shaped the substantive agenda by authoring seed statements, adding new statements and making patterns of agreement, disagreement and uncertainty made visible through voting.Our endeavors represent one of the the first instances of applying PD to a venue that critically interrogates the societal impacts of AI, fostering a niche in which critical scholars are free to voice their concerns. Finally, this work advances large-scale PD theory by providing an effective case study of a co-design paradigm that can readily scale temporally and epistemologically.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 2 minor

Summary. The manuscript reports a participatory design (PD) process for ACM FAccT conference governance that combined an in-person CRAFT session (where participants authored seed statements on conference issues) with an asynchronous Polis poll (for voting on agreement, disagreement, and uncertainty) and the subsequent synthesis of inputs into a governance report for FAccT leadership. The authors position the effort as one of the first applications of PD to a venue focused on the societal impacts of AI and as an advance in large-scale PD theory via a scalable co-design paradigm.

Significance. If the participation data, voting patterns, and synthesis process were fully documented and shown to be representative, the work could provide a practical case study for community-driven reflexive governance in interdisciplinary AI ethics conferences. The hybrid CRAFT+Polis approach demonstrates a concrete way to surface critical scholar concerns in such venues and offers a model that other conferences might adapt. The manuscript's descriptive framing of the process is a strength, but the absence of supporting evidence limits its ability to advance PD theory or substantiate claims of effectiveness and scalability.

major comments (3)
  1. [Abstract] Abstract: The claim that the work 'advances large-scale PD theory by providing an effective case study of a co-design paradigm that can readily scale temporally and epistemologically' is unsupported because the manuscript reports no participant counts, response rates, demographics, voting breakdowns, or agreement patterns from the CRAFT session or Polis poll.
  2. [Process description] Process description (CRAFT and Polis sections): No analysis of self-selection bias, affiliation/field breakdowns, or representativeness is provided, which is load-bearing for the central claim that the process produced representative input for FAccT governance and constitutes a 'first instance' of PD in this venue.
  3. [Synthesis and outcomes] Synthesis and outcomes section: The manuscript does not detail how raw Polis votes and authored statements were aggregated or filtered into the final governance report, preventing assessment of whether synthesis artifacts affected the actionability of the results.
minor comments (2)
  1. [Abstract] Abstract contains a repeated word: 'one of the the first instances' should read 'one of the first instances'.
  2. [Results] The manuscript would be strengthened by including at least summary tables or figures of key Polis voting results and the main themes in the synthesized report.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their constructive and detailed review, which identifies key areas for strengthening the manuscript's claims and transparency. We address each major comment point by point below, with plans for revisions where appropriate to improve the paper without misrepresenting the work.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The claim that the work 'advances large-scale PD theory by providing an effective case study of a co-design paradigm that can readily scale temporally and epistemologically' is unsupported because the manuscript reports no participant counts, response rates, demographics, voting breakdowns, or agreement patterns from the CRAFT session or Polis poll.

    Authors: We agree that the abstract's phrasing overstates the theoretical contribution without supporting quantitative details, as the manuscript is primarily a descriptive case study of the hybrid PD process applied to FAccT governance. The claim was intended to highlight the methodological innovation in combining in-person CRAFT authoring with asynchronous Polis voting for scalability, but we acknowledge the lack of reported metrics limits substantiation. We will revise the abstract to qualify or remove the 'advances large-scale PD theory' language, focusing instead on the practical demonstration of the approach and its potential for temporal and epistemological scaling in similar communities. revision: yes

  2. Referee: [Process description] Process description (CRAFT and Polis sections): No analysis of self-selection bias, affiliation/field breakdowns, or representativeness is provided, which is load-bearing for the central claim that the process produced representative input for FAccT governance and constitutes a 'first instance' of PD in this venue.

    Authors: The referee is correct that the manuscript lacks explicit analysis of self-selection bias or demographic breakdowns, which weakens claims of representativeness. Participation was open to all FAccT 2023 attendees via the CRAFT session and subsequent Polis poll, with no targeted sampling or demographic collection beyond conference registration data. We will add a dedicated limitations paragraph in the process description section discussing voluntary participation, potential biases toward engaged scholars, and the distinction between statistical representativeness and the goal of surfacing diverse critical perspectives. For the 'first instance' claim, we will provide additional context from a review of prior FAccT programs to support it as one of the earliest documented PD efforts for venue governance, while noting it is not exhaustive. revision: partial

  3. Referee: [Synthesis and outcomes] Synthesis and outcomes section: The manuscript does not detail how raw Polis votes and authored statements were aggregated or filtered into the final governance report, preventing assessment of whether synthesis artifacts affected the actionability of the results.

    Authors: We accept this critique and will substantially expand the synthesis section. The process involved thematic grouping of high-consensus Polis statements (defined as those with >60% agreement and <20% disagreement where possible), integration of CRAFT-authored seeds that aligned with poll themes, and filtering for actionability by prioritizing items relevant to conference governance (e.g., program structure, inclusion). We will add a step-by-step account of this aggregation, including any manual curation steps and how uncertainty votes were handled, to enable evaluation of the results' robustness. revision: yes

Circularity Check

0 steps flagged

No circularity in this descriptive case study

full rationale

This is a qualitative case study reporting a participatory design process (CRAFT session + Polis poll + report synthesis) for FAccT governance. The paper contains no equations, fitted parameters, predictions, uniqueness theorems, or derivation chains. The central claim of advancing large-scale PD theory rests on describing the reported process and outcomes rather than any self-referential reduction or self-citation load-bearing step. No steps match the enumerated circularity patterns; the work is self-contained as a descriptive account.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on standard assumptions from HCI and design research about the value and scalability of participatory methods; no free parameters, new entities, or ad-hoc inventions are introduced in the abstract.

axioms (1)
  • domain assumption Participatory design can effectively capture diverse community inputs and produce scalable governance insights for academic conferences.
    Invoked throughout the description of the process and its claimed benefits.

pith-pipeline@v0.9.0 · 5487 in / 1323 out tokens · 35980 ms · 2026-05-10T07:30:33.106365+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

23 extracted references · 23 canonical work pages

  1. [1]

    Arnstein

    Sherry R. Arnstein. 1969. A Ladder Of Citizen Participation.Journal of the American Institute of Planners35, 4 (July 1969), 216–224. doi:10.1080/01944366908977225

    work page doi:10.1080/01944366908977225 1969

  2. [2]

    Abeba Birhane, William Isaac, Vinodkumar Prabhakaran, Mark Diaz, Madeleine Clare Elish, Iason Gabriel, and Shakir Mohamed. 2022. Power to the people? Opportunities and challenges for participatory AI. InProceedings of the 2nd ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization. 1–8

    work page 2022

  3. [3]

    Susanne Bødker, Christian Dindler, and Ole Sejer Iversen. 2017. Tying Knots: Participatory Infrastructuring at Work.Computer Supported Cooperative Work (CSCW)26, 1–2 (2017), 245–273. doi:10.1007/s10606-017-9268-y

    work page doi:10.1007/s10606-017-9268-y 2017

  4. [4]

    Susanne Bødker, Christian Dindler, Ole S Iversen, and Rachel C Smith. 2022. What is participatory design? InParticipatory design. Springer, 5–13

    work page 2022

  5. [5]

    Claus Bossen, Reem Talhouk, and John Vines. 2025. Scaling Participatory Design. InRoutledge International Handbook of Contemporary Participatory Design, Rachel Charlotte Smith, Daria Loi, Heike Winschiers-Theophilus, Liesbeth Huybrechts, and Jesper Simonsen (Eds.). Routledge, London. doi:10.4324/9781003334330-3

    work page doi:10.4324/9781003334330-3 2025

  6. [6]

    2020.Design justice: Community-led practices to build the worlds we need

    Sasha Costanza-Chock. 2020.Design justice: Community-led practices to build the worlds we need. MIT press

    work page 2020

  7. [7]

    Fernando Delgado, Stephen Yang, Michael Madaio, and Qian Yang. 2023. The participatory turn in ai design: Theoretical foundations and the current state of practice. InProceedings of the 3rd ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization. 1–23

    work page 2023

  8. [8]

    Saffron Huang, Divya Siddarth, Liane Lovitt, Thomas I Liao, Esin Durmus, Alex Tamkin, and Deep Ganguli. 2024. Collective constitutional ai: Aligning a language model with public input. InProceedings of the 2024 ACM Conference on Fairness, Accountability, and Transparency. 1395–1417

    work page 2024

  9. [9]

    Liesbeth Huybrechts, Henric Benesch, and Jon Geib. 2017. Institutioning: Participatory design, co-design and the public realm.CoDesign 13, 3 (2017), 148–159

    work page 2017

  10. [10]

    Liesbeth Huybrechts, Maurizio Teli, Joanna Saad-Sulonen, and Mariacristina Sciannamblo. 2025. Collectives and Participatory Design. In Routledge International Handbook of Contemporary Participatory Design, Rachel Charlotte Smith, Daria Loi, Heike Winschiers-Theophilus, Liesbeth Huybrechts, and Jesper Simonsen (Eds.). Routledge, London. doi:10.4324/9781003334330-4

    work page doi:10.4324/9781003334330-4 2025

  11. [11]

    Helena Karasti. 2014. Infrastructuring in participatory design. InProceedings of the 13th Participatory Design Conference: Research Papers-Volume 1. 141–150

    work page 2014

  12. [12]

    Benjamin Laufer, Sameer Jain, A Feder Cooper, Jon Kleinberg, and Hoda Heidari. 2022. Four years of FAccT: A reflexive, mixed- methods analysis of research contributions, shortcomings, and future prospects. InProceedings of the 2022 ACM conference on fairness, accountability, and transparency. 401–426

    work page 2022

  13. [13]

    Organizers Of Queerinai, Anaelia Ovalle, Arjun Subramonian, Ashwin Singh, Claas Voelcker, Danica J Sutherland, Davide Locatelli, Eva Breznik, Filip Klubicka, Hang Yuan, et al. 2023. Queer in AI: A case study in community-led participatory AI. InProceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency. 1882–1895

    work page 2023

  14. [14]

    Lars Kristian Roland, Terje Aksel Sanner, Johan Ivar Sæbø, and Eric Monteiro. 2017. P for Platform. Architectures of large-scale participatory design.Scandinavian Journal of Information Systems29, 2 (2017), 1

    work page 2017

  15. [15]

    Hope Schroeder, Akshansh Pareek, and Solon Barocas. 2025. Disclosure without Engagement: An Empirical Review of Positionality Statements at FAccT. InProceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency. 1195–1210. Taking Stock at FAccT FAccT ’26, June 25–28, 2026, Montréal, Canada

    work page 2025

  16. [16]

    Ali Akbar Septiandri, Marios Constantinides, Mohammad Tahaei, and Daniele Quercia. 2023. WEIRD FAccTs: How western, educated, industrialized, rich, and democratic is FAccT?. InProceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency. 160–171

    work page 2023

  17. [17]

    Mona Sloane, Emanuel Moss, Olaitan Awomolo, and Laura Forlano. 2022. Participation is not a design fix for machine learning. In Proceedings of the 2nd ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization. 1–6

    work page 2022

  18. [18]

    Christopher Small, Michael Bjorkegren, Timo Erkkilä, Lynette Shaw, and Colin Megill. 2021. Polis: Scaling deliberation by mapping high dimensional opinion spaces.Recerca: revista de pensament i anàlisi26, 2 (2021)

    work page 2021

  19. [19]

    Christopher T Small, Ivan Vendrov, Esin Durmus, Hadjar Homaei, Elizabeth Barry, Julien Cornebise, Ted Suzman, Deep Ganguli, and Colin Megill. 2023. Opportunities and risks of LLMs for scalable deliberation with Polis.arXiv preprint arXiv:2306.11932(2023)

    work page arXiv 2023

  20. [20]

    Ownership, Not Just Happy Talk

    Emily Tseng, Meg Young, Marianne Aubin Le Quéré, Aimee Rinehart, and Harini Suresh. 2025. " Ownership, Not Just Happy Talk": Co-Designing a Participatory Large Language Model for Journalism. InProceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency. 3119–3130

    work page 2025

  21. [21]

    Daly, and Aditi Verma

    Peter Wacnik, Shanna R. Daly, and Aditi Verma. 2025. Participatory Design: A Systematic Review and Insights for Future Practice. Design Science11 (2025), e21. doi:10.1017/dsj.2025.10009

    work page doi:10.1017/dsj.2025.10009 2025

  22. [22]

    Meg Young, Upol Ehsan, Ranjit Singh, Emnet Tafesse, Michele Gilman, Christina Harrington, and Jacob Metcalf. 2024. Participation versus scale: Tensions in the practical demands on participatory AI.First Monday(2024)

    work page 2024

  23. [23]

    non-academics

    Øivind Klungseth Zahlsen, Elena Parmiggiani, and Yngve Dahl. 2022. Challenges of Scaling Participatory Design: A Systematic Literature Review. InProceedings of the 34th Australian Conference on Human-Computer Interaction (OzCHI ’22). ACM, New York, NY. doi:10.1145/3572921.3572924 FAccT ’26, June 25–28, 2026, Montréal, Canada Shiran Dudy*, Jan Simson*, and...

    work page doi:10.1145/3572921.3572924 2022