Engaged AI Governance: Addressing the Last Mile Challenge Through Internal Expert Collaboration

Orestis Papakyriakopoulos; Simon Jarvers

arxiv: 2604.21554 · v1 · submitted 2026-04-23 · 💻 cs.AI

Engaged AI Governance: Addressing the Last Mile Challenge Through Internal Expert Collaboration

Simon Jarvers , Orestis Papakyriakopoulos This is my paper

Pith reviewed 2026-05-09 21:11 UTC · model grok-4.3

classification 💻 cs.AI

keywords AI governanceEU AI Actlast mile challengepractitioner perceptionsregulatory complianceaction researchsoftware developmentinternal collaboration

0 comments

The pith

Internal expert collaboration turns EU AI Act requirements into shared development priorities rather than external box-ticking.

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

The paper investigates the practical difficulty of turning high-level AI regulations into concrete software development steps. It uses insider research inside one AI startup to build and test a pipeline that pulls requirements from legal text, involves practitioners in assessment and idea generation, and lets the team collectively rank what to implement. The study finds three recurring ways developers view these requirements: some align directly with their existing goals, some match work already underway, and some register as disconnected administrative tasks. From this the authors conclude that rules are handled genuinely only when teams see how following them improves the system and protects users; otherwise compliance stays superficial. Expert collaboration inside the team is presented as the mechanism that can shift governance from imposed obligation to collective ownership.

Core claim

Through insider action research the authors construct a legal-text-to-action pipeline that extracts EU AI Act requirements, engages practitioners in assessment and ideation, and prioritizes actions through collective evaluation. Analysis of the resulting discussions identifies three perception patterns: convergence, in which compliance matches development priorities; existing practice, in which current work already meets the requirement; and disconnection, in which the requirement feels like administrative overhead. The paper claims that this distinction determines whether governance is treated genuinely or performatively, and that internal expert collaboration supplies the practical route,

What carries the argument

The legal-text-to-action pipeline, which extracts requirements from legal text, engages practitioners in assessment and ideation, and prioritizes implementation through collective evaluation.

Load-bearing premise

The three perception patterns observed inside one AI startup reflect how practitioners generally interpret regulatory requirements and that internal expert collaboration will reliably produce shared ownership in other organizations and regulatory settings.

What would settle it

A study that tracks the same requirements in multiple teams, some using internal expert collaboration and some not, and measures whether the non-collaborative teams treat a higher share of requirements as disconnected overhead and produce lower-quality compliance artifacts.

Figures

Figures reproduced from arXiv: 2604.21554 by Orestis Papakyriakopoulos, Simon Jarvers.

**Figure 1.** Figure 1: AI Governance Levels and the “Last Mile” Challenge. AI governance operates at three levels: industry (regulation and [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗

**Figure 2.** Figure 2: Action Research Cycle mapped to paper structure. Adapted from Coghlan and Brannick [9]. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** Figure 3: From legal text to action: A translation pipeline for operationalizing EU AI Act requirements. The collaborative [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

**Figure 4.** Figure 4: Impact-effort mapping of the 14 workshop-generated strategies. Participants discussed each strategy until reaching [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗

read the original abstract

Under the EU AI Act, translating AI governance requirements into software development practice remains challenging. While AI governance frameworks exist at industry and organizational levels, empirical evidence of team-level implementation is scarce. We address this "Last Mile" Challenge through insider action research embedded within an AI startup. We present a legal-text-to-action pipeline that translates EU AI Act requirements into actionable strategies through internal expert collaboration by extracting requirements from legal text, engaging practitioners in assessment and ideation, and prioritizing implementation through collective evaluation. Our analysis reveals three patterns in how practitioners perceive regulatory requirements: convergence (compliance aligns with development priorities), existing practice (current work already satisfies requirements), and disconnection (requirements perceived as administrative overhead). Based on these patterns, we discuss when governance might be treated genuinely or performatively. Practitioners prioritize requirements that serve end-users or their own development needs, but view verification-oriented requirements as box-ticking exercises. This distinction suggests a translation challenge: regulatory requirements risk superficial treatment unless practitioners understand how compliance serves system quality and user protection. Expert collaboration offers a practical mechanism for transforming governance from external imposition to shared ownership and making previously invisible governance work visible and collective.

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.

Desk Editor's Note private letter to a colleague

Single-case insider study in one startup gives a plausible pipeline and three perception patterns for EU AI Act implementation but supplies almost no methodological detail.

read the letter

This paper reports insider action research from one AI startup that built a legal-text-to-action pipeline and surfaced three patterns in how practitioners view regulatory requirements: convergence with existing priorities, overlap with current practice, and disconnection as overhead. The central suggestion is that internal expert collaboration can turn external rules into shared ownership instead of box-ticking. That framing of the last-mile problem is the main thing worth taking away. The pipeline steps themselves (extract requirements, engage the team in assessment and ideation, then prioritize collectively) are described clearly enough to be tried by others, and the observation that teams treat user-protection rules more seriously than verification rules matches what many engineers report in practice. Those elements give the work some immediate utility for people actually shipping AI systems under the EU AI Act. The soft spot is the evidence base. Everything rests on observations from a single embedded team with no numbers on participants, no description of how the patterns were coded or validated, and no discussion of how the researchers managed their own insider position. Without those details the patterns could be idiosyncratic to that startup's size, culture, or the specific requirements they examined. The paper does not attempt cross-case comparison or falsification steps that would strengthen the claim that the pipeline reliably converts governance into genuine work. Readers who already work on AI governance implementation or who study regulatory translation in organizations will find the practical framing useful and may borrow the pipeline idea. Academic readers looking for robust qualitative evidence will see the gaps immediately. The work deserves peer review because the topic is timely and the approach is grounded in real practice, but any referee would need to press on methods, limitations, and scope before the claims could be treated as more than a single-case illustration.

Referee Report

2 major / 2 minor

Summary. The paper claims that insider action research in one AI startup reveals a 'Last Mile Challenge' in translating EU AI Act requirements into practice. It introduces a legal-text-to-action pipeline (extract requirements, engage practitioners in assessment/ideation, prioritize via collective evaluation) and identifies three practitioner perception patterns—convergence (compliance aligns with priorities), existing practice (current work suffices), and disconnection (requirements seen as overhead)—arguing these indicate a translation challenge best addressed by internal expert collaboration to shift from performative to genuine governance and make invisible work visible and collective.

Significance. If the patterns and pipeline hold, the work offers practical empirical insight into how AI teams perceive and prioritize regulatory requirements, highlighting risks of superficial compliance when requirements are not linked to system quality or user protection. The action research approach is a strength for surfacing real-world dynamics and proposing a collaborative mechanism. However, the single-case scope limits significance to hypothesis generation for broader governance research rather than validated general mechanisms.

major comments (2)

[Methods / Empirical Study] The empirical foundation for the three patterns and the translation challenge claim lacks reported details on participant numbers, data collection methods (e.g., workshops or interviews), analysis procedures (e.g., how patterns were coded), study duration, and bias mitigation for the insider position. This is load-bearing because the patterns directly support the central argument that the pipeline fosters shared ownership rather than performative compliance.
[Findings and Discussion] The generalization that internal expert collaboration transforms external requirements into shared ownership is based solely on observations from one EU AI Act-focused startup. No cross-organizational comparisons, inter-rater reliability checks, or falsification attempts are described, so the patterns risk being idiosyncratic to team size, culture, or researcher position rather than indicating a reliable broader mechanism.

minor comments (2)

[Abstract] The abstract supplies no information on study scale or methods, which would help readers immediately assess the empirical basis for the patterns.
[Pipeline Description] Clarify notation for the pipeline stages and ensure all figures or tables (if present) explicitly link observed patterns to specific pipeline steps.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback, which identifies key areas for improving the transparency of our methods and the appropriate scoping of our claims. We address each major comment below, indicating where revisions will be made to the manuscript.

read point-by-point responses

Referee: [Methods / Empirical Study] The empirical foundation for the three patterns and the translation challenge claim lacks reported details on participant numbers, data collection methods (e.g., workshops or interviews), analysis procedures (e.g., how patterns were coded), study duration, and bias mitigation for the insider position. This is load-bearing because the patterns directly support the central argument that the pipeline fosters shared ownership rather than performative compliance.

Authors: We agree that the current manuscript would be strengthened by more explicit methodological reporting. In the revised version, we will add a dedicated Methods subsection that specifies: the number of practitioners who participated in the workshops, assessments, and ideation sessions; the data collection approaches (including workshop formats, follow-up discussions, and observation periods); the qualitative analysis process (thematic coding steps used to derive the three perception patterns); the overall duration of the embedded insider research; and the specific steps taken to mitigate insider researcher bias, such as triangulation across data sources and member checking with participants. These additions will make the empirical basis for the patterns and the pipeline more transparent and reproducible. revision: yes
Referee: [Findings and Discussion] The generalization that internal expert collaboration transforms external requirements into shared ownership is based solely on observations from one EU AI Act-focused startup. No cross-organizational comparisons, inter-rater reliability checks, or falsification attempts are described, so the patterns risk being idiosyncratic to team size, culture, or researcher position rather than indicating a reliable broader mechanism.

Authors: We accept that the study is limited to a single organizational case and do not claim the patterns represent validated general mechanisms. The work is positioned as insider action research that surfaces context-specific insights and generates hypotheses for further investigation, consistent with the significance assessment provided. In revision, we will update the Discussion to more explicitly bound the claims, noting potential influences of team size, culture, and the researcher's embedded position, and we will call for future multi-case studies to examine broader applicability. The collaborative analysis process among the authors will be described in greater detail; formal inter-rater reliability metrics are not standard for this interpretive approach but can be supplemented with additional process transparency. We cannot incorporate cross-organizational data or falsification tests without a new study design, but the contribution remains the detailed account of the legal-text-to-action pipeline and observed patterns from deep embedded practice. revision: partial

Circularity Check

0 steps flagged

No circularity: empirical patterns derived from primary observations

full rationale

The paper presents a qualitative insider action research study in one AI startup. Its central outputs—the three practitioner perception patterns (convergence, existing practice, disconnection) and the proposed legal-text-to-action pipeline—are direct interpretations of data collected during the embedded study. No equations, fitted parameters, or self-referential definitions appear; the patterns are not constructed by renaming prior results or by load-bearing self-citations that presuppose the target claim. The derivation chain is therefore self-contained as standard empirical analysis rather than tautological.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on standard assumptions of qualitative action research and the relevance of the EU AI Act to startup development practice; no quantitative free parameters or newly postulated entities are introduced.

axioms (1)

domain assumption Insider action research embedded in an organization can generate valid insights into how practitioners perceive and respond to external regulatory requirements.
The entire study is built on this embedded research approach described in the abstract.

pith-pipeline@v0.9.0 · 5499 in / 1422 out tokens · 51101 ms · 2026-05-09T21:11:15.380403+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

38 extracted references · 38 canonical work pages

[1]

Ali, Angèle Christin, Andrew Smart, and Riitta Katila

Sanna J. Ali, Angèle Christin, Andrew Smart, and Riitta Katila. 2023. Walking the Walk of AI Ethics: Organizational Challenges and the Individualization of Risk among Ethics Entrepreneurs. InProceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency(Chicago, IL, USA)(FAccT ’23). Association for Computing Machinery, New York, NY, ...

work page doi:10.1145/3593013 2023
[2]

Peter M. Asaro. 2000. Transforming society by transforming technology: the science and politics of participatory design.Accounting, Management and Information Technologies10, 4 (2000), 257–290. doi:10.1016/S0959-8022(00)00004-7

work page doi:10.1016/s0959-8022(00)00004-7 2000
[3]

Amna Batool, Didar Zowghi, and Muneera Bano. 2025. AI governance: a systematic literature review.AI and Ethics5, 3 (2025), 3265–3279. doi:10.1007/s43681-024-00653-w

work page doi:10.1007/s43681-024-00653-w 2025
[4]

Elettra Bietti. 2020. From ethics washing to ethics bashing: a view on tech ethics from within moral philosophy. InProceedings of the 2020 Conference on Fairness, Accountability, and Transparency(Barcelona, Spain)(FAT* ’20). Association for Computing Machinery, New York, NY, USA, 210–219. doi:10.1145/3351095.3372860

work page doi:10.1145/3351095.3372860 2020
[5]

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(Arlington, VA, USA)(EAAMO ’22). Association for Com...

work page doi:10.1145/3551624.3555290 2022
[6]

Teemu Birkstedt, Matti Minkkinen, Anushree Tandon, and Matti Mäntymäki. 2023. AI governance: themes, knowledge gaps and future agendas.Internet Research33, 7 (06 2023), 133–167. arXiv:https://www.emerald.com/intr/article-pdf/33/7/133/1214024/intr-01-2022- 0042.pdf doi:10.1108/INTR-01-2022-0042

work page doi:10.1108/intr-01-2022-0042 2023
[7]

Edyta Bogucka, Marios Constantinides, Sanja Šćepanović, and Daniele Quercia. 2024. Co-designing an AI Impact Assessment Report Template with AI Practitioners and AI Compliance Experts.Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society7, 1 (Oct. 2024), 168–180. doi:10.1609/aies.v7i1.31627

work page doi:10.1609/aies.v7i1.31627 2024
[8]

Mark Anthony Camilleri. 2024. Artificial intelligence governance: Ethical considerations and implications for social responsibility. Expert Systems41, 7 (2024), e13406. arXiv:https://onlinelibrary.wiley.com/doi/pdf/10.1111/exsy.13406 doi:10.1111/exsy.13406

work page doi:10.1111/exsy.13406 2024
[9]

2019.Doing Action Research in Your Own Organization

David Coghlan. 2019.Doing Action Research in Your Own Organization. SAGE Publications Ltd, London :. http://digital.casalini.it/ 9781526481719

work page 2019
[10]

Keeley Crockett, Edwin Colyer, Luciano Gerber, and Annabel Latham. 2023. Building Trustworthy AI Solutions: A Case for Practical Solutions for Small Businesses.IEEE Transactions on Artificial Intelligence4, 4 (2023), 778–791. doi:10.1109/TAI.2021.3137091

work page doi:10.1109/tai.2021.3137091 2023
[11]

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(Boston, MA, USA)(EAAMO ’23). Association for Computing Machinery, New York, NY, USA, Ar...

work page doi:10.1145/3617694.3623261 2023
[12]

Wesley Hanwen Deng, Nur Yildirim, Monica Chang, Motahhare Eslami, Kenneth Holstein, and Michael Madaio. 2023. Investigating Practices and Opportunities for Cross-functional Collaboration around AI Fairness in Industry Practice. InProceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency(Chicago, IL, USA)(FAccT ’23). Association ...

work page doi:10.1145/3593013.3594037 2023
[13]

Virginia Dignum. 2023. Responsible Artificial Intelligence: Recommendations and Lessons Learned. InResponsible AI in Africa: Challenges and Opportunities, Damian Okaibedi Eke, Kutoma Wakunuma, and Simisola Akintoye (Eds.). Springer International Publishing, Cham, 195–214. doi:10.1007/978-3-031-08215-3_9

work page doi:10.1007/978-3-031-08215-3_9 2023
[14]

Thilo Hagendorff. 2020. The ethics of AI ethics: An evaluation of guidelines.Minds and machines30, 1 (2020), 99–120. doi:10.1007/s11023- 020-09517-8

work page doi:10.1007/s11023- 2020
[15]

Tomasz Hollanek, Yulu Pi, Cosimo Fiorini, Virginia Vignali, Dorian Peters, and Eleanor Drage. 2025. A Toolkit for Compliance, a Toolkit for Justice: Drawing on Cross-sectoral Expertise to Develop a Pro-justice EU AI Act Toolkit. InProceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’25). Association for Computing Ma...

work page doi:10.1145/3715275.3732078 2025
[16]

Aspen Hopkins and Serena Booth. 2021. Machine Learning Practices Outside Big Tech: How Resource Constraints Challenge Responsible Development. InProceedings of the 2021 AAAI/ACM Conference on AI, Ethics, and Society(Virtual Event, USA)(AIES ’21). Association for FAccT ’26, June 25–28, 2026, Montreal, QC, Canada Jarvers & Papakyriakopoulos Computing Machin...

work page doi:10.1145/3461702.3462527 2021
[17]

Anna Jobin, Marcello Ienca, and Effy Vayena. 2019. The global landscape of AI ethics guidelines.Nature machine intelligence1, 9 (2019), 389–399. doi:10.1038/s42256-019-0088-2

work page doi:10.1038/s42256-019-0088-2 2019
[18]

Emma Kallina, Thomas Bohné, and Jatinder Singh. 2025. Stakeholder Participation for Responsible AI Development: Disconnects Between Guidance and Current Practice. InProceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’25). Association for Computing Machinery, New York, NY, USA, 1060–1079. doi:10.1145/3715275.3732069

work page doi:10.1145/3715275.3732069 2025
[19]

Emma Kallina and Jatinder Singh. 2024. Stakeholder Involvement for Responsible AI Development: A Process Framework. InProceedings of the 4th ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization(San Luis Potosi, Mexico)(EAAMO ’24). Association for Computing Machinery, New York, NY, USA, Article 1, 14 pages. doi:10.1145/3689904.3694698

work page doi:10.1145/3689904.3694698 2024
[20]

Fiona Koh, Kathrin Grosse, and Giovanni Apruzzese. 2024. Voices from the Frontline: Revealing the AI Practitioners’ viewpoint on the European AI Act. InProceedings of the Annual Hawaii International Conference on System Sciences. Hawaii International Conference on System Sciences, Maui, Hawaii, USA, 1870–1879. doi:10.24251/HICSS.2024.235

work page doi:10.24251/hicss.2024.235 2024
[21]

Who", “What

Blaine Kuehnert, Rachel Kim, Jodi Forlizzi, and Hoda Heidari. 2025. The “Who", “What", and “How" of Responsible AI Governance: A Systematic Review and Meta-Analysis of (Actor, Stage)-Specific Tools. InProceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’25). Association for Computing Machinery, New York, NY, USA, 29...

work page arXiv 2025
[22]

Qinghua Lu, Liming Zhu, Xiwei Xu, Jon Whittle, Didar Zowghi, and Aurelie Jacquet. 2024. Responsible AI Pattern Catalogue: A Collection of Best Practices for AI Governance and Engineering.ACM Comput. Surv.56, 7, Article 173 (April 2024), 35 pages. doi:10.1145/3626234

work page doi:10.1145/3626234 2024
[23]

Robert MacIntosh, Marc Bonnet, and David Coghlan. 2007. Insider action research: opportunities and challenges.Management Research News30, 5 (05 2007), 335–343. arXiv:https://www.emerald.com/mrr/article-pdf/30/5/335/2055095/01409170710746337.pdf doi:10.1108/01409170710746337

work page doi:10.1108/01409170710746337 2007
[24]

Mariano Méndez-Suárez, Virginia Simón-Moya, and Javier Muñoz-de Prat. 2023. Do current regulations prevent unethical AI practices? Journal of Competitiveness15, 3 (2023), 207. doi:10.7441/joc.2023.03.11

work page doi:10.7441/joc.2023.03.11 2023
[25]

Brent Mittelstadt. 2019. Principles alone cannot guarantee ethical AI.Nature machine intelligence1, 11 (2019), 501–507. doi:10.1038/s42256- 019-0114-4

work page doi:10.1038/s42256- 2019
[26]

Jakob Mökander, Maria Axente, Federico Casolari, and Luciano Floridi. 2022. Conformity assessments and post-market monitoring: a guide to the role of auditing in the proposed European AI regulation.Minds and Machines32, 2 (2022), 241–268. doi:10.1007/s11023-021-09577-4

work page doi:10.1007/s11023-021-09577-4 2022
[27]

Jakob Mökander and Luciano Floridi. 2023. Operationalising AI governance through ethics-based auditing: an industry case study.AI and Ethics3, 2 (2023), 451–468. doi:10.1007/s43681-022-00171-7

work page doi:10.1007/s43681-022-00171-7 2023
[28]

Luke Munn. 2023. The uselessness of AI ethics.AI and Ethics3, 3 (2023), 869–877. doi:10.1007/s43681-022-00209-w

work page doi:10.1007/s43681-022-00209-w 2023
[29]

Nadia Nahar, Shurui Zhou, Grace Lewis, and Christian Kästner. 2022. Collaboration challenges in building ML-enabled systems: communication, documentation, engineering, and process. InProceedings of the 44th International Conference on Software Engineering (Pittsburgh, Pennsylvania)(ICSE ’22). Association for Computing Machinery, New York, NY, USA, 413–425...

work page doi:10.1145/3510003.3510209 2022
[30]

Emmanouil Papagiannidis, Patrick Mikalef, and Kieran Conboy. 2025. Responsible artificial intelligence governance: A review and research framework.The Journal of Strategic Information Systems34, 2 (2025), 101885. doi:10.1016/j.jsis.2024.101885

work page doi:10.1016/j.jsis.2024.101885 2025
[31]

Petar Radanliev, Omar Santos, Alistair Brandon-Jones, and Adam Joinson. 2024. Ethics and responsible AI deployment.Frontiers in Artificial Intelligence7 (2024), 1377011. doi:10.3389/frai.2024.1377011

work page doi:10.3389/frai.2024.1377011 2024
[32]

Bogdana Rakova, Jingying Yang, Henriette Cramer, and Rumman Chowdhury. 2021. Where Responsible AI meets Reality: Practitioner Perspectives on Enablers for Shifting Organizational Practices.Proc. ACM Hum.-Comput. Interact.5, CSCW1, Article 7 (April 2021), 23 pages. doi:10.1145/3449081

work page doi:10.1145/3449081 2021
[33]

Lorenn P Ruster and Jenny L Davis. 2025. The Gaps that Never Were: Reconsidering Responsible AI’s Principle-Practice Problem. In Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’25). Association for Computing Machinery, New York, NY, USA, 350–360. doi:10.1145/3715275.3732024

work page doi:10.1145/3715275.3732024 2025
[34]

Morgan Klaus Scheuerman. 2024. In the Walled Garden: Challenges and Opportunities for Research on the Practices of the AI Tech Industry. InProceedings of the 2024 ACM Conference on Fairness, Accountability, and Transparency(Rio de Janeiro, Brazil)(FAccT ’24). Association for Computing Machinery, New York, NY, USA, 456–466. doi:10.1145/3630106.3658918

work page doi:10.1145/3630106.3658918 2024
[35]

Ben Shneiderman. 2020. Bridging the Gap Between Ethics and Practice: Guidelines for Reliable, Safe, and Trustworthy Human-centered AI Systems.ACM Trans. Interact. Intell. Syst.10, 4, Article 26 (Oct. 2020), 31 pages. doi:10.1145/3419764

work page doi:10.1145/3419764 2020
[36]

Societal biases in language generation: Progress and challenges

Mona Sloane and Janina Zakrzewski. 2022. German AI Start-Ups and “AI Ethics”: Using A Social Practice Lens for Assessing and Implementing Socio-Technical Innovation. InProceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency (Seoul, Republic of Korea)(FAccT ’22). Association for Computing Machinery, New York, NY, USA, 935–947. ...

work page doi:10.1145/3531146.3533156 2022
[37]

Chiara Ullstein, Simon Jarvers, Michel Hohendanner, Orestis Papakyriakopoulos, and Jens Grossklags. 2025. Participatory AI and the EU AI Act.Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society8, 3 (Oct. 2025), 2550–2562. doi:10.1609/aies.v8i3.36737

work page doi:10.1609/aies.v8i3.36737 2025
[38]

Previously when customers reported errors, we relied on their descriptions which were often incomplete. Now we can trace exactly what happened

Rui-Jie Yew, Bill Marino, and Suresh Venkatasubramanian. 2025. Red Teaming AI Policy: A Taxonomy of Avoision and the EU AI Act. In Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’25). Association for Computing Machinery, Engaged AI Governance Through Internal Expert Collaboration FAccT ’26, June 25–28, 2026, Mo...

work page doi:10.1145/3715275.3732028 2025

[1] [1]

Ali, Angèle Christin, Andrew Smart, and Riitta Katila

Sanna J. Ali, Angèle Christin, Andrew Smart, and Riitta Katila. 2023. Walking the Walk of AI Ethics: Organizational Challenges and the Individualization of Risk among Ethics Entrepreneurs. InProceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency(Chicago, IL, USA)(FAccT ’23). Association for Computing Machinery, New York, NY, ...

work page doi:10.1145/3593013 2023

[2] [2]

Peter M. Asaro. 2000. Transforming society by transforming technology: the science and politics of participatory design.Accounting, Management and Information Technologies10, 4 (2000), 257–290. doi:10.1016/S0959-8022(00)00004-7

work page doi:10.1016/s0959-8022(00)00004-7 2000

[3] [3]

Amna Batool, Didar Zowghi, and Muneera Bano. 2025. AI governance: a systematic literature review.AI and Ethics5, 3 (2025), 3265–3279. doi:10.1007/s43681-024-00653-w

work page doi:10.1007/s43681-024-00653-w 2025

[4] [4]

Elettra Bietti. 2020. From ethics washing to ethics bashing: a view on tech ethics from within moral philosophy. InProceedings of the 2020 Conference on Fairness, Accountability, and Transparency(Barcelona, Spain)(FAT* ’20). Association for Computing Machinery, New York, NY, USA, 210–219. doi:10.1145/3351095.3372860

work page doi:10.1145/3351095.3372860 2020

[5] [5]

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(Arlington, VA, USA)(EAAMO ’22). Association for Com...

work page doi:10.1145/3551624.3555290 2022

[6] [6]

Teemu Birkstedt, Matti Minkkinen, Anushree Tandon, and Matti Mäntymäki. 2023. AI governance: themes, knowledge gaps and future agendas.Internet Research33, 7 (06 2023), 133–167. arXiv:https://www.emerald.com/intr/article-pdf/33/7/133/1214024/intr-01-2022- 0042.pdf doi:10.1108/INTR-01-2022-0042

work page doi:10.1108/intr-01-2022-0042 2023

[7] [7]

Edyta Bogucka, Marios Constantinides, Sanja Šćepanović, and Daniele Quercia. 2024. Co-designing an AI Impact Assessment Report Template with AI Practitioners and AI Compliance Experts.Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society7, 1 (Oct. 2024), 168–180. doi:10.1609/aies.v7i1.31627

work page doi:10.1609/aies.v7i1.31627 2024

[8] [8]

Mark Anthony Camilleri. 2024. Artificial intelligence governance: Ethical considerations and implications for social responsibility. Expert Systems41, 7 (2024), e13406. arXiv:https://onlinelibrary.wiley.com/doi/pdf/10.1111/exsy.13406 doi:10.1111/exsy.13406

work page doi:10.1111/exsy.13406 2024

[9] [9]

2019.Doing Action Research in Your Own Organization

David Coghlan. 2019.Doing Action Research in Your Own Organization. SAGE Publications Ltd, London :. http://digital.casalini.it/ 9781526481719

work page 2019

[10] [10]

Keeley Crockett, Edwin Colyer, Luciano Gerber, and Annabel Latham. 2023. Building Trustworthy AI Solutions: A Case for Practical Solutions for Small Businesses.IEEE Transactions on Artificial Intelligence4, 4 (2023), 778–791. doi:10.1109/TAI.2021.3137091

work page doi:10.1109/tai.2021.3137091 2023

[11] [11]

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(Boston, MA, USA)(EAAMO ’23). Association for Computing Machinery, New York, NY, USA, Ar...

work page doi:10.1145/3617694.3623261 2023

[12] [12]

Wesley Hanwen Deng, Nur Yildirim, Monica Chang, Motahhare Eslami, Kenneth Holstein, and Michael Madaio. 2023. Investigating Practices and Opportunities for Cross-functional Collaboration around AI Fairness in Industry Practice. InProceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency(Chicago, IL, USA)(FAccT ’23). Association ...

work page doi:10.1145/3593013.3594037 2023

[13] [13]

Virginia Dignum. 2023. Responsible Artificial Intelligence: Recommendations and Lessons Learned. InResponsible AI in Africa: Challenges and Opportunities, Damian Okaibedi Eke, Kutoma Wakunuma, and Simisola Akintoye (Eds.). Springer International Publishing, Cham, 195–214. doi:10.1007/978-3-031-08215-3_9

work page doi:10.1007/978-3-031-08215-3_9 2023

[14] [14]

Thilo Hagendorff. 2020. The ethics of AI ethics: An evaluation of guidelines.Minds and machines30, 1 (2020), 99–120. doi:10.1007/s11023- 020-09517-8

work page doi:10.1007/s11023- 2020

[15] [15]

Tomasz Hollanek, Yulu Pi, Cosimo Fiorini, Virginia Vignali, Dorian Peters, and Eleanor Drage. 2025. A Toolkit for Compliance, a Toolkit for Justice: Drawing on Cross-sectoral Expertise to Develop a Pro-justice EU AI Act Toolkit. InProceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’25). Association for Computing Ma...

work page doi:10.1145/3715275.3732078 2025

[16] [16]

Aspen Hopkins and Serena Booth. 2021. Machine Learning Practices Outside Big Tech: How Resource Constraints Challenge Responsible Development. InProceedings of the 2021 AAAI/ACM Conference on AI, Ethics, and Society(Virtual Event, USA)(AIES ’21). Association for FAccT ’26, June 25–28, 2026, Montreal, QC, Canada Jarvers & Papakyriakopoulos Computing Machin...

work page doi:10.1145/3461702.3462527 2021

[17] [17]

Anna Jobin, Marcello Ienca, and Effy Vayena. 2019. The global landscape of AI ethics guidelines.Nature machine intelligence1, 9 (2019), 389–399. doi:10.1038/s42256-019-0088-2

work page doi:10.1038/s42256-019-0088-2 2019

[18] [18]

Emma Kallina, Thomas Bohné, and Jatinder Singh. 2025. Stakeholder Participation for Responsible AI Development: Disconnects Between Guidance and Current Practice. InProceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’25). Association for Computing Machinery, New York, NY, USA, 1060–1079. doi:10.1145/3715275.3732069

work page doi:10.1145/3715275.3732069 2025

[19] [19]

Emma Kallina and Jatinder Singh. 2024. Stakeholder Involvement for Responsible AI Development: A Process Framework. InProceedings of the 4th ACM Conference on Equity and Access in Algorithms, Mechanisms, and Optimization(San Luis Potosi, Mexico)(EAAMO ’24). Association for Computing Machinery, New York, NY, USA, Article 1, 14 pages. doi:10.1145/3689904.3694698

work page doi:10.1145/3689904.3694698 2024

[20] [20]

Fiona Koh, Kathrin Grosse, and Giovanni Apruzzese. 2024. Voices from the Frontline: Revealing the AI Practitioners’ viewpoint on the European AI Act. InProceedings of the Annual Hawaii International Conference on System Sciences. Hawaii International Conference on System Sciences, Maui, Hawaii, USA, 1870–1879. doi:10.24251/HICSS.2024.235

work page doi:10.24251/hicss.2024.235 2024

[21] [21]

Who", “What

Blaine Kuehnert, Rachel Kim, Jodi Forlizzi, and Hoda Heidari. 2025. The “Who", “What", and “How" of Responsible AI Governance: A Systematic Review and Meta-Analysis of (Actor, Stage)-Specific Tools. InProceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’25). Association for Computing Machinery, New York, NY, USA, 29...

work page arXiv 2025

[22] [22]

Qinghua Lu, Liming Zhu, Xiwei Xu, Jon Whittle, Didar Zowghi, and Aurelie Jacquet. 2024. Responsible AI Pattern Catalogue: A Collection of Best Practices for AI Governance and Engineering.ACM Comput. Surv.56, 7, Article 173 (April 2024), 35 pages. doi:10.1145/3626234

work page doi:10.1145/3626234 2024

[23] [23]

Robert MacIntosh, Marc Bonnet, and David Coghlan. 2007. Insider action research: opportunities and challenges.Management Research News30, 5 (05 2007), 335–343. arXiv:https://www.emerald.com/mrr/article-pdf/30/5/335/2055095/01409170710746337.pdf doi:10.1108/01409170710746337

work page doi:10.1108/01409170710746337 2007

[24] [24]

Mariano Méndez-Suárez, Virginia Simón-Moya, and Javier Muñoz-de Prat. 2023. Do current regulations prevent unethical AI practices? Journal of Competitiveness15, 3 (2023), 207. doi:10.7441/joc.2023.03.11

work page doi:10.7441/joc.2023.03.11 2023

[25] [25]

Brent Mittelstadt. 2019. Principles alone cannot guarantee ethical AI.Nature machine intelligence1, 11 (2019), 501–507. doi:10.1038/s42256- 019-0114-4

work page doi:10.1038/s42256- 2019

[26] [26]

Jakob Mökander, Maria Axente, Federico Casolari, and Luciano Floridi. 2022. Conformity assessments and post-market monitoring: a guide to the role of auditing in the proposed European AI regulation.Minds and Machines32, 2 (2022), 241–268. doi:10.1007/s11023-021-09577-4

work page doi:10.1007/s11023-021-09577-4 2022

[27] [27]

Jakob Mökander and Luciano Floridi. 2023. Operationalising AI governance through ethics-based auditing: an industry case study.AI and Ethics3, 2 (2023), 451–468. doi:10.1007/s43681-022-00171-7

work page doi:10.1007/s43681-022-00171-7 2023

[28] [28]

Luke Munn. 2023. The uselessness of AI ethics.AI and Ethics3, 3 (2023), 869–877. doi:10.1007/s43681-022-00209-w

work page doi:10.1007/s43681-022-00209-w 2023

[29] [29]

Nadia Nahar, Shurui Zhou, Grace Lewis, and Christian Kästner. 2022. Collaboration challenges in building ML-enabled systems: communication, documentation, engineering, and process. InProceedings of the 44th International Conference on Software Engineering (Pittsburgh, Pennsylvania)(ICSE ’22). Association for Computing Machinery, New York, NY, USA, 413–425...

work page doi:10.1145/3510003.3510209 2022

[30] [30]

Emmanouil Papagiannidis, Patrick Mikalef, and Kieran Conboy. 2025. Responsible artificial intelligence governance: A review and research framework.The Journal of Strategic Information Systems34, 2 (2025), 101885. doi:10.1016/j.jsis.2024.101885

work page doi:10.1016/j.jsis.2024.101885 2025

[31] [31]

Petar Radanliev, Omar Santos, Alistair Brandon-Jones, and Adam Joinson. 2024. Ethics and responsible AI deployment.Frontiers in Artificial Intelligence7 (2024), 1377011. doi:10.3389/frai.2024.1377011

work page doi:10.3389/frai.2024.1377011 2024

[32] [32]

Bogdana Rakova, Jingying Yang, Henriette Cramer, and Rumman Chowdhury. 2021. Where Responsible AI meets Reality: Practitioner Perspectives on Enablers for Shifting Organizational Practices.Proc. ACM Hum.-Comput. Interact.5, CSCW1, Article 7 (April 2021), 23 pages. doi:10.1145/3449081

work page doi:10.1145/3449081 2021

[33] [33]

Lorenn P Ruster and Jenny L Davis. 2025. The Gaps that Never Were: Reconsidering Responsible AI’s Principle-Practice Problem. In Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’25). Association for Computing Machinery, New York, NY, USA, 350–360. doi:10.1145/3715275.3732024

work page doi:10.1145/3715275.3732024 2025

[34] [34]

Morgan Klaus Scheuerman. 2024. In the Walled Garden: Challenges and Opportunities for Research on the Practices of the AI Tech Industry. InProceedings of the 2024 ACM Conference on Fairness, Accountability, and Transparency(Rio de Janeiro, Brazil)(FAccT ’24). Association for Computing Machinery, New York, NY, USA, 456–466. doi:10.1145/3630106.3658918

work page doi:10.1145/3630106.3658918 2024

[35] [35]

Ben Shneiderman. 2020. Bridging the Gap Between Ethics and Practice: Guidelines for Reliable, Safe, and Trustworthy Human-centered AI Systems.ACM Trans. Interact. Intell. Syst.10, 4, Article 26 (Oct. 2020), 31 pages. doi:10.1145/3419764

work page doi:10.1145/3419764 2020

[36] [36]

Societal biases in language generation: Progress and challenges

Mona Sloane and Janina Zakrzewski. 2022. German AI Start-Ups and “AI Ethics”: Using A Social Practice Lens for Assessing and Implementing Socio-Technical Innovation. InProceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency (Seoul, Republic of Korea)(FAccT ’22). Association for Computing Machinery, New York, NY, USA, 935–947. ...

work page doi:10.1145/3531146.3533156 2022

[37] [37]

Chiara Ullstein, Simon Jarvers, Michel Hohendanner, Orestis Papakyriakopoulos, and Jens Grossklags. 2025. Participatory AI and the EU AI Act.Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society8, 3 (Oct. 2025), 2550–2562. doi:10.1609/aies.v8i3.36737

work page doi:10.1609/aies.v8i3.36737 2025

[38] [38]

Previously when customers reported errors, we relied on their descriptions which were often incomplete. Now we can trace exactly what happened

Rui-Jie Yew, Bill Marino, and Suresh Venkatasubramanian. 2025. Red Teaming AI Policy: A Taxonomy of Avoision and the EU AI Act. In Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency (FAccT ’25). Association for Computing Machinery, Engaged AI Governance Through Internal Expert Collaboration FAccT ’26, June 25–28, 2026, Mo...

work page doi:10.1145/3715275.3732028 2025