Global AI Governance: Where the Challenge is the Solution- An Interdisciplinary, Multilateral, and Vertically Coordinated Approach

Hong Shen; Huixin Zhong; Rostam J. Neuwirth; Thao Do; Ynagliu Jie

REVIEW 3 major objections 5 minor 1 cited by

Global AI Governance: Where the Challenge is the Solution- An Interdisciplinary, Multilateral, and Vertically Coordinated Approach

T0 review · 3 major / 5 minor · reviewed 2026-08-08 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This paper claims that the gridlock of global AI governance—disciplinary fragmentation, stalled multilateralism, and the gap between policy and practice—can be turned into a five-phase framework that assigns law, civil society, and…

desk verdict A coherent, clearly written synthesis of three expert views into a five-phase AI governance roadmap, but the phase-action mapping is asserted rather than derived, so the 'concrete pathway' claim is not supported. read the letter →

arxiv 2503.04766 v1 pith:RLUV35RQ submitted 2025-02-12 cs.CY

Huixin Zhong , Thao Do , Ynagliu Jie , Rostam J. Neuwirth , Hong Shen This is my paper · ORCID

classification cs.CY

keywords globalAIgovernanceinterdisciplinarydialoguemultilateralcoordinationverticalparadoxicalpolicyexperimentationparticipatorydesignfive-phaseframework

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper argues that the obstacles to global AI governance—fragmented disciplines, stalled multilateral talks, and the disconnect between top-down policy and grassroots use—are not just problems to be overcome but materials to design with. Using a structured interdisciplinary dialogue among one legal expert, one Global South civil-society practitioner, and one human-computer interaction researcher, the authors extract four principles on which all three agree: dynamism, experimentation, inclusivity, and paradoxical governance. These principles are then translated into concrete instruments—regulatory sandboxes, policy pilots, and participatory design workshops—and arranged into a five-phase, time-sequential roadmap from foundation building through continuous evolution. A sympathetic reader should care because the framework offers a concrete path from abstract global norms to implementable actions, with each phase assigning a distinct task to law, civil society, and user-centered design. The study's reach depends on whether three purposefully chosen voices provide enough diversity to anchor a global framework, a limitation the authors acknowledge.

What carries the argument

The central object is the paper's 'synergistic multilevel framework' built from four principles and five phases, with the study's sampling design—'minimum necessary diversity'—as the load-bearing input. The four principles (dynamism, experimentation, inclusivity, paradoxical governance) are the consensual common ground of three experts selected to span an authority–justice–technology tension triangle: legal authority at the macro level, distributive justice for the Global South at the meso level, and technological democratization at the micro level. The framework's work is performed by pairing each principle with a concrete instrument—legal sandboxes and cross-border contract templates (law), policy pilots and adversarial negotiation platforms (civil society), and user-driven audits and participatory co-design workshops (HCI)—and sequencing those instruments into five phases: foundation building, experimental verification, collaborative optimization, global adaptation, and continuous evolution. The mechanism that carries the argument is the complementarity of the three disciplines: law supplies institutional extensibility, the civil-society perspective supplies political legitimacy and equity pressure, and HCI supplies user feedback and local adaptation, so that each phase converts a governance tension into a division of labor.

What would settle it

Run the same structured dialogue with a different triad—say an economist, a national data-protection regulator, and a frontline public-service worker, or a panel drawn from several regions of the Global South—and compare the resulting principles. If the four principles shift materially across triads, the claim that three minimally diverse voices suffice for a global framework is falsified. A second check: implement a legal sandbox with a policy pilot and user audits in one jurisdiction and see whether the predicted feedback loop changes the rules; if it stalls, the framework's dynamism mechanism fails.

Watch

Extended reading notes

Core claim

At the paper's core is the claim that a minimally diverse panel—one expert each from law, Global South civil society, and human-computer interaction—can, through a structured dialogue, surface a shared set of governance principles that the broader policy world has failed to agree on. The authors report that all three experts converge on four principles: governance must be dynamic (adapting as technology changes), experimental (learning through pilots and tests), inclusive (drawing in users and marginalized regions), and paradoxical (treating conflicting objectives as tensions to manage rather than problems to eliminate). The paper's five-phase framework then operationalizes these principles in a specific sequence: build the legal and cultural foundation, run controlled experiments, optimize collaboratively, adapt globally, and keep evolving via a monitoring-analysis-iteration loop. Each phase assigns complementary roles—legal sandboxes from the legal perspective, regional and South-South policy pilots from the NGO perspective, and user audits and co-design from the HCI perspective—so that abstract principles become concrete management measures. If the authors are right, the same structured dialogue that revealed the principles is itself a reusable governance mechanism: the challenge of disagreement is converted into the solution of a coordinated framework.

Load-bearing premise

The entire framework rests on the assumption that three purposefully selected experts—one in law, one in Global South civil society, and one in human-computer interaction—represent enough of the world's diversity that a global AI governance framework built from their shared principles will generalize to all stakeholders.

Editorial extensions

If this is right

International bodies could replace the 'regulatory race' narrative with a repeatable process: convene a minimally diverse panel, extract shared principles, and run the five phases in sequence.
The same template can be transferred to other global governance domains, such as climate change, because the tension between global norms and local practice is generic—the paper's final section says exactly this.
The framework gives users, civil society, and Global South voices a formal, phase-specific implementation role rather than a consultative one, changing how national AI strategies would be constructed.
Each principle becomes auditable through a concrete instrument: dynamism through legal sandboxes and feedback loops, experimentation through policy pilots, inclusivity through participatory design, and paradoxical governance through adversarial negotiation mechanisms.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The paper leaves implicit that 'the challenge is the solution' generalizes beyond AI: any governance deadlock can be treated as a design input, which suggests an adversarial-negotiation mechanism could be tested against consensus-based international bodies.
A stress test the authors do not run: scale the dialogue from three to dozens of stakeholders and check whether the four principles stay stable or proliferate; instability would show the minimal-diversity premise is too thin.
An actionable criterion follows from the framework: a proposed governance action counts as on-framework only if it advances at least one of dynamism, experimentation, inclusivity, or paradox management—useful for auditing future AI policy proposals.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Read this one if you want a quick picture of where AI governance proposals currently sit. The paper's actual contribution is a five-phase roadmap—foundation building, experimental verification, collaborative optimization, global adaptation, continuous evolution—derived from a three-expert panel at the UN Science Summit. That specific synthesis of IIS with legal/NGO/HCI perspectives is not something I've seen in the cited literature. The writing is clear, the OSF materials are provided, and Section 9 is honest about the small sample and limited diversity.

What it does well: the four principles (dynamism, experimentation, inclusivity, paradoxical governance) are a reasonable summary of common ground across the three talks, and the interdisciplinary comparison is a legitimate way to surface complementarities. The framework is readable and could serve as a discussion scaffold for international organizations.

The soft spots are substantial, though. The empirical base is three talks. There is no intercoder reliability, no member checking, no comparison against existing governance frameworks. More importantly, the stress-test is right: there is no decision rule connecting the four principles to the five phases. Section 6 assigns specific actions (legal sandboxes, policy pilots, participatory workshops) to phases, but nothing in the data shows why those actions belong in those phases or why that ordering is necessary. An alternative grouping would be just as consistent. The abstract's claim of a 'novel and concrete pathway' is therefore overstated. The paper gives a plausible menu of governance tools, not a validated pathway.

The 'first' and 'seminal' language in Sections 8 and 10 should be toned down; it is not supported by the evidence or by a systematic check of prior frameworks.

Bottom line: this is a thoughtful proposal with a weak empirical foundation. It deserves peer review because the topic is important and the framework is testable, but a serious referee should demand much stronger evidence—more experts, a codebook, independent validation, and a real comparison with existing frameworks. I wouldn't cite it as an empirical result, but I'd use it in a reading group to discuss qualitative methods in governance research.

Referee Report

3 major / 5 minor

Summary. The paper reports a qualitative study conducted at the UN Science Summit in which three experts from law, a Global South NGO, and HCI each gave a presentation and participated in a panel discussion. Through thematic analysis and interdisciplinary comparative analysis, the authors identify four principles of global AI governance—dynamism, experimentation, inclusivity, and paradoxical governance—and then construct a five-phase, time-sequential framework (foundation building, experimental verification, collaborative optimization, global adaptation, continuous evolution) that assigns concrete actions such as legal sandboxes, policy pilots, and participatory design workshops to each phase. The central claim is that this framework offers a novel and concrete pathway for interdisciplinary, multilateral, and vertically coordinated AI governance.

Significance. If the proposed framework were rigorously derived and validated, it would provide a useful practical scaffolding for combining top-down and bottom-up governance mechanisms, and the use of IIS as an organizing methodology for a UN-based dialogue is a constructive idea. The paper has several strengths: it openly provides an OSF link to transcripts and thematic analysis, it explicitly acknowledges limitations in sample size and diversity, and it attempts to translate abstract governance principles into concrete, phase-specific actions rather than stopping at general recommendations. However, the evidentiary basis is thin—three expert inputs—and the chain from data to principles to phases is not transparently documented. The significance of the contribution depends on whether the phase-action mapping can be shown to be more than a plausible narrative, which the manuscript does not currently establish.

major comments (3)

[§3.2 and §6] The central claim that the five-phase framework is a 'concrete pathway' rests on the unexplained step from four principles to five phases and to the specific phase-action assignments. The paper reports that two researchers independently coded transcripts and that sub-themes were clustered into meta-propositions, but it provides no inter-coder reliability statistic, no codebook, and no decision rule that connects the 15/14/11 sub-themes to the four principles or the principles to the actions placed in each phase of Section 6. As it stands, the same actions could plausibly be reordered—for example, policy pilots in Phase 2 and participatory design workshops in Phase 3 could be swapped—without contradicting any of the four principles. The 'concrete pathway' claim is therefore not yet falsifiable from the reported evidence.
[§3.1 and §9] The sampling design rests on the assertion that three purposefully selected experts satisfy 'minimum necessary diversity' for a global AI governance framework, but the paper provides no justification for this assertion beyond invoking an authority-justice-technology triangle. The authors' own Section 9 acknowledges that the small sample size and limited diversity restrict generalizability. This is not a peripheral caveat: the framework's claimed globality and multilateral applicability depend on the representativeness of these three voices. A single NGO practitioner and a single HCI researcher cannot credibly stand for the Global South or for user communities, and the paper should either provide additional evidence for the sufficiency of this diversity or substantially soften the global claims.
[§8 and §10] The contributions section describes the framework as 'the first integrative framework' and 'seminal,' and the conclusion calls it 'the first global AI governance framework' of its kind. These claims require a systematic comparison against existing frameworks, such as the UNESCO Recommendation on the Ethics of AI, the OECD AI Principles, the EU AI Act, and the UN High-Level Advisory Body's proposals. No such comparison is provided. The novelty claim should be supported by a literature gap analysis or reformulated as a context-specific proposal based on three expert perspectives rather than as a historically first framework.

minor comments (5)

[§1] The Introduction lists the five-phase process as 'experimental verification, collaborative optimization, global adaptation, and continuous evolution,' omitting 'foundation building' that appears in the abstract and in Section 6. This inconsistency should be corrected.
[§2.1.2] The phrase 'In this respires' appears to be a typographical error for 'In this respect.'
[§5] The headings 'Complimentary' should read 'Complementary' in the four subsections of Section 5.
[§3.2] The paper should clarify what the OSF repository contains beyond transcripts and thematic analysis; specifically, whether the full codebook, the coding of sub-themes, and the phase-mapping materials are available for inspection. This would strengthen the reproducibility of the analysis.
[References] Several reference entries are incomplete or contain placeholders, for example references [2] and [3] include 'Accessed: [date you accessed the website]' rather than an actual access date. The reference formatting should be made consistent.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the framework is an inductive qualitative synthesis rather than a fitted prediction or self-citation chain.

full rationale

The paper's derivation chain is an inductive qualitative synthesis, not a fitted model. The four principles are presented as the outcome of thematic coding and interdisciplinary comparison of the three expert talks (§3.2, §5), and the five-phase framework is explicitly constructed from those same principles and the experts' complementary proposals (§6). This is theory building from the data, not a prediction that is then 'confirmed' by the same data. No parameter is fitted and then renamed as a result; no uniqueness theorem from the authors' prior work is invoked as a forced choice; and the authors' cited prior work (e.g., Shen et al. on user-driven auditing, Neuwirth on legal paradox) is external published evidence used as background for the expert themes, not as the justification for the framework's validity. The absence of an explicit decision rule linking the four principles to the five phases, and the lack of inter-coder reliability statistics, are reproducibility and internal-validity weaknesses, not circularity: an alternative phase ordering would be a different synthesis, not a contradiction of a fitted prediction. Section 9 explicitly acknowledges the generalizability limitation, confirming that the paper does not claim an independent empirical test of the framework. The abstract's compressed phrasing, 'Drawing on the common ground of the experts: dynamism, experimentation, inclusivity, and paradoxical governance, this study ... identifies four core principles,' places the conclusion before the method, but the method section does derive the themes from the transcripts, so this is summarization rather than a definitional reduction. Overall, the central claim is a proposed pathway, not a result that reduces by construction to its own inputs.

Assumptions & free parameters 0 free parameters · 3 assumptions · 1 invented entities

The central framework rests on three qualitative assumptions: the sufficiency of the three-expert sample, the objectivity of the thematic coding, and the suitability of IIS as an organizing lens. None of these is independently justified. There are no numeric free parameters. The framework itself is an invented conceptual construct with no external validation.

assumptions (3)

ad hoc to paper The three selected experts (legal, NGO, HCI) constitute 'minimum necessary diversity' sufficient to represent global AI governance perspectives.
Invoked in Section 3.1; no empirical justification that three domains suffice for global coverage.
domain assumption Thematic analysis by two researchers and subsequent interdisciplinary comparison reliably extracts objective themes from the talks.
Section 3.2; intercoder reliability and validity of the synthesis are not reported.
domain assumption Integration and Implementation Science (IIS) is an appropriate meta-framework for global AI governance.
Section 3; the choice of IIS as lens is asserted, not derived or justified against alternatives.

invented entities (1)

Synergistic multilevel framework (five phases: foundation building, experimental verification, collaborative optimization, global adaptation, continuous evolution)
purpose: Operationalize the four principles into a time-sequential governance process.
Introduced in Section 6; no external empirical validation or falsifiable handle is provided beyond the authors' proposal.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Global AI Governance: Where the Challenge is the Solution- An Interdisciplinary, Multilateral, and Vertically Coordinated Approach." pith.science (2026). https://pith.science/paper/RLUV35RQ

@misc{pith2026250304766,
  author       = {Pith},
  title        = {Pith review of: Global AI Governance: Where the Challenge is the Solution- An Interdisciplinary, Multilateral, and Vertically Coordinated Approach},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RLUV35RQ}},
  note         = {Machine review of arXiv:2503.04766}
}

read the original abstract

Current global AI governance frameworks struggle with fragmented disciplinary collaboration, ineffective multilateral coordination, and disconnects between policy design and grassroots implementation. This study, guided by Integration and Implementation Science (IIS) initiated a structured interdisciplinary dialogue at the UN Science Summit, convening legal, NGO, and HCI experts to tackle those challenges. Drawing on the common ground of the experts: dynamism, experimentation, inclusivity, and paradoxical governance, this study, through thematic analysis and interdisciplinary comparison analysis, identifies four core principles of global AI governance. Furthermore, we translate these abstract principles into concrete action plans leveraging the distinct yet complementary perspectives of each discipline. These principles and action plans are then integrated into a five-phase, time-sequential framework including foundation building, experimental verification, collaborative optimization, global adaptation, and continuous evolution phases. This multilevel framework offers a novel and concrete pathway toward establishing interdisciplinary, multilateral, and vertically coordinated AI governance, transforming global AI governance challenges into opportunities for political actions.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

SCITUS: A Multi-Jurisdictional Framework for Adapting NIST AI RMF to the Canadian Regulatory Context
cs.CY 2026-07 conditional novelty 6.0 of 10

SCITUS maps 127 Canadian federal and provincial AI requirements onto 57 NIST-aligned controls and argues this unified approach beats jurisdiction-by-jurisdiction compliance.

Reference graph

Works this paper leans on

48 extracted references · 46 canonical work pages · cited by 1 Pith paper

[1]

Ai act, 2024

European Commission. Ai act, 2024. 27th, Dec, 2024

work page 2024
[2]

Blueprint for an ai bill of rights: Making automated systems work for the american people,

The White House. Blueprint for an ai bill of rights: Making automated systems work for the american people,

work page
[3]

Ethical norms for a new generation of artificial intelligence (english translation by center for security and emerging technology), 2021

The National New Generation Artificial Intelligence Governance Specialist Committee. Ethical norms for a new generation of artificial intelligence (english translation by center for security and emerging technology), 2021. Accessed: [date you accessed the website]

work page 2021
[4]

Ai regulation: a pro-innovation approach - white paper, mar

Innovation Department for Science and Technology. Ai regulation: a pro-innovation approach - white paper, mar

work page
[5]

Digital empires: The global battle to regulate technology

Anu Bradford. Digital empires: The global battle to regulate technology. Oxford University Press, 2023

work page 2023
[6]

From a ‘race to ai’to a ‘race to ai regulation’: regulatory competition for artificial intelligence

Nathalie A Smuha. From a ‘race to ai’to a ‘race to ai regulation’: regulatory competition for artificial intelligence. Law, Innovation and Technology, 13(1):57–84, 2021

work page 2021
[7]

Artificial intelligence between national strategies and the global regulatory race: analytical routes for an international and comparative reappraisal

Pasquot Polido and Fabricio Bertini. Artificial intelligence between national strategies and the global regulatory race: analytical routes for an international and comparative reappraisal. Rev. Faculdade Direito Universidade Federal Minas Gerais, 76:229, 2020

work page 2020
[8]

The global institutional governance of ai: A four -dimensional perspective

Rostam J Neuwirth. The global institutional governance of ai: A four -dimensional perspective. International Journal of Digital Law and Governance, 1(1):113–153, 2024

work page 2024

Show all 48 references

[9]

Against the new space race: Global ai competition and cooperation for people

Inga Ulnicane. Against the new space race: Global ai competition and cooperation for people. AI & SOCIETY, 38(2):681–683, 2023

2023
[10]

Recommendation on the ethics of artificial intelligence, 2022

UNESCO. Recommendation on the ethics of artificial intelligence, 2022

2022
[11]

Establishing the rules for building trustworthy ai

Luciano Floridi. Establishing the rules for building trustworthy ai. Ethics, Governance, and Policies in Artificial Intelligence, pages 41–45, 2021

2021
[12]

Recommendation of the council on consumer protection in e-commerce

OECD. Recommendation of the council on consumer protection in e-commerce. OECD Legal Instruments, 2024. Accessed: 2024-12-27

2024
[13]

Governing AI for Humanity: Final Report, 2024

United Nations AI Advisory Body. Governing AI for Humanity: Final Report, 2024. Accessed: 2024-12-30

2024
[14]

Governing AI for Humanity: Interim Report, 2024

United Nations AI Advisory Body. Governing AI for Humanity: Interim Report, 2024. Accessed: 2024-12-30

2024
[15]

Data warfare and creating a global legal and regulatory landscape: Challenges and solutions

Varda Mone, Sadikov Maksudboy Abdulajonovich, Ammar Younas, and Sailaja Petikam. Data warfare and creating a global legal and regulatory landscape: Challenges and solutions. International Journal of Legal Information, October 2024

2024
[16]

The middle-out approach: assessing models of legal governance in data protection, artificial intelligence, and the web of data

Ugo Pagallo, Pompeu Cas anovas, and Robert Madelin. The middle-out approach: assessing models of legal governance in data protection, artificial intelligence, and the web of data. The Theory and Practice of Legislation, 7(1), January 2019

2019
[17]

Integration and implementation sciences: building a new specialization

Gabriele Bammer et al. Integration and implementation sciences: building a new specialization. Ecol. Soc , 10(2):95–107, 2003

2003
[18]

Maximizing team synergy in ai-related interdisciplinary groups: an interdisciplinary-by-design iterative methodology

Piercosma Bisconti, Davide Orsitto, Federica Fedorczyk, Fabio Brau, Marianna Capasso, Lorenzo De Marinis, HÃ¼seyin Eken, Federica Merenda, Mirko Forti, Marco Pacini, and Claudia Schettini. Maximizing team synergy in ai-related interdisciplinary groups: an interdisciplinary-by-...

2022
[19]

Regulating ai: Applying insights from behavioural economics and psychology to the application of article 5 of the eu ai act

Huixin Zhong, Eamonn O’Neill, and Janina A Hoffmann. Regulating ai: Applying insights from behavioural economics and psychology to the application of article 5 of the eu ai act. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 20001–20009, 2024

2024
[20]

The eu artificial intelligence act: Regulating subliminal ai systems

Rostam J Neuwirth. The eu artificial intelligence act: Regulating subliminal ai systems. 2023

2023
[21]

Why and how is the power of big tech increasing in the policy process? the case of generative ai

Shaleen Khanal, Hongzhou Zhang, and Araz Taeihagh. Why and how is the power of big tech increasing in the policy process? the case of generative ai. Policy and Society, March 2024

2024
[22]

The politics of method in the human sciences: Positivism and its epistemological others

George Steinmetz. The politics of method in the human sciences: Positivism and its epistemological others. American Historical Review, 110(5):1638, 2005

2005
[23]

The atlas of ai: Power, politics, and the planetary costs of artificial intelligence, 2021

Kate Crawford. The atlas of ai: Power, politics, and the planetary costs of artificial intelligence, 2021

2021
[24]

Using thematic analysis in psychology

Virginia Braun and Victoria Clarke. Using thematic analysis in psychology. Qualitative research in psychology, 3(2):77–101, 2006

2006
[25]

Interdisciplinary research: Process and theory

Allen F Repko and Rick Szostak. Interdisciplinary research: Process and theory. Sage publications, 2020. Global AI Governance: Where the Challenge is the Solution- An Interdisciplinary, Multilateral, and Vertically Coordinated Approach 11

2020
[26]

Law in the time of oxymora: A synaesthesia of language, logic and law

Rostam J Neuwirth. Law in the time of oxymora: A synaesthesia of language, logic and law. Routledge, 2018

2018
[27]

The rational design of international institutions

Barbara Koremenos, Charles Lipson, and Duncan Snidal. The rational design of international institutions. International Organization, 55(4), 2001

2001
[28]

Governance in 21st century: Global governance, 1995

J Rosenau. Governance in 21st century: Global governance, 1995

1995
[29]

Paradoxical tensions in the implementation of digital security governance: toward an ambidextrous approach to governing digital security

Stef Schinagl, Abbas Shahim, and Svetlana Khapova. Paradoxical tensions in the implementation of digital security governance: toward an ambidextrous approach to governing digital security. Computers & Security , 122:102903, 2022

2022
[30]

Artificial intelligence regulation: a framework for governance

Patricia Gomes RÃªgo de Almeida, Carlos Denner dos Santos, and Josivania Silva Farias. Artificial intelligence regulation: a framework for governance. Ethics and Information Technology, 23(3), April 2021

2021
[31]

Everyday algorithm auditing: Understanding the power of everyday users in surfacing harmful algorithmic behaviors

Hong Shen, Alicia DeVos, Motahhare Eslami, and Kenneth Holstein. Everyday algorithm auditing: Understanding the power of everyday users in surfacing harmful algorithmic behaviors. Proceedings of the ACM on Human - Computer Interaction, 5(CSCW2):1–29, 2021

2021
[32]

Toward user-driven algorithm auditing: Investigating users’ strategies for uncovering harmful algorithmic behavior

Alicia DeVos, Aditi Dhabalia, Hong Shen, Kenneth Holstein, and Motahhare Eslami. Toward user-driven algorithm auditing: Investigating users’ strategies for uncovering harmful algorithmic behavior. In Proceedings of the 2022 CHI conference on human factors in computing systems,...

2022
[33]

User-driven value alignment: Understanding users’ perceptions and strategies for addressing biased and discriminatory statements in ai companions

Xianzhe Fan, Qing Xiao, Xuhui Zhou, Jiaxin Pei, Maarten Sap, Zhicong Lu, and Hong Shen. User-driven value alignment: Understanding users’ perceptions and strategies for addressing biased and discriminatory statements in ai companions. arXiv preprint arXiv:2409.00862, 2024

2024 arXiv
[34]

Investigating what factors influence users’ rating of harmful algorithmic bias and discrimination

Sara Kingsley, Jiayin Zhi, Wesley Hanwen Deng, Jaimie Lee, Sizhe Zhang, Motahhare Eslami, Kenneth Holstein, Jason I Hong, Tianshi Li, and Hong Shen. Investigating what factors influence users’ rating of harmful algorithmic bias and discrimination. In Proceedings of the AAAI Co...

2024
[35]

Rena Li, Sara Kingsley, Chelsea Fan, Proteeti Sinha, Nora Wai, Jaimie Lee, Hong Shen, Motahhare Esl ami, and Jason Hong. Participation and division of labor in user -driven algorithm audits: How do everyday users work together to surface algorithmic harms? In Proceedings of th...

2023
[36]

public (s) -in-the-loop

Hong Shen, Ángel Alexander Cabrera, Adam Perer, and Jason Hong. " public (s) -in-the-loop": Facilitating deliberation of algorithmic decisions in contentious public policy domains. arXiv preprint arXiv:2204.10814, 2022

2022 arXiv
[37]

The model card authoring toolkit: Toward community-centered, deliberation-driven ai design

Hong Shen, Leijie Wang, Wesley H Deng, Ciell Brusse, Ronald Velgersdijk, and Haiyi Zhu. The model card authoring toolkit: Toward community-centered, deliberation-driven ai design. In Proceedings of the 2022 ACM Conference on Fairness, Accountability, and Transparency, pages 44...

2022
[38]

Value cards: An educational toolkit for teaching social impacts of machine learning through deliberation

Hong Shen, Wesley H Deng, Aditi Chattopadhyay, Zhiwei Steven Wu, Xu Wang, and Haiyi Zhu. Value cards: An educational toolkit for teaching social impacts of machine learning through deliberation. In Proceedings of the 2021 ACM conference on fairness, accountability, and transpa...

2021
[39]

Designing alterna- tive representations of confusion matrices to support non-expert public understanding of algorithm performance

Hong Shen, Haojian Jin, Ángel Alexander Cabrera, Adam Perer, Haiyi Zhu, and Jason I Hong. Designing alterna- tive representations of confusion matrices to support non-expert public understanding of algorithm performance. Proceedings of the ACM on Human-Computer Interaction, 4(...

2020
[40]

Ai failure cards: Understanding and supporting grassroots efforts to mitigate ai failures in homeless services

Ningjing Tang, Jiayin Zhi, Tzu -Sheng Kuo, Calla Kainaroi, Jeremy J Northup, Kenneth Holstein, Haiyi Zhu, Hoda Heidari, and Hong Shen. Ai failure cards: Understanding and supporting grassroots efforts to mitigate ai failures in homeless services. In The 2024 ACM Conference on ...

2024
[41]

Understand- ing frontline workers’ and unhoused individuals’ perspectives on ai used in homeless services

Tzu-Sheng Kuo, Hong Shen, Jisoo Geum, Nev Jones, Jason I Hong, Haiyi Zhu, and Kenneth Holstein. Understand- ing frontline workers’ and unhoused individuals’ perspectives on ai used in homeless services. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Sy...

2023
[42]

Interdisciplinarity: History, theory, and practice, 1990

Julie Thompson Klein. Interdisciplinarity: History, theory, and practice, 1990

1990
[43]

The constitution of society: Outline of the theory of structuration

Anthony Giddens. The constitution of society: Outline of the theory of structuration. Polity, 1984

1984
[44]

Exploration and exploitation in organizational learning

James G March. Exploration and exploitation in organizational learning. Organization science, 2(1):71–87, 1991

1991
[45]

Re-thinking science: Knowledge and the public in an age of uncertainty

Helga Nowotny. Re-thinking science: Knowledge and the public in an age of uncertainty. Polity, 2001

2001
[46]

Between facts and norms

J Habermas. Between facts and norms. MIT Press Cambridge MA, 1996

1996
[47]

Toward a theory of paradox: A dynamic equilibrium model of organizing

Wendy K Smith and Marianne W Lewis. Toward a theory of paradox: A dynamic equilibrium model of organizing. Academy of management Review, 36(2):381–403, 2011

2011
[2023]

Accessed: [date you accessed the website]

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.

[1] [1]

Ai act, 2024

European Commission. Ai act, 2024. 27th, Dec, 2024

work page 2024

[2] [2]

Blueprint for an ai bill of rights: Making automated systems work for the american people,

The White House. Blueprint for an ai bill of rights: Making automated systems work for the american people,

work page

[3] [3]

Ethical norms for a new generation of artificial intelligence (english translation by center for security and emerging technology), 2021

The National New Generation Artificial Intelligence Governance Specialist Committee. Ethical norms for a new generation of artificial intelligence (english translation by center for security and emerging technology), 2021. Accessed: [date you accessed the website]

work page 2021

[4] [4]

Ai regulation: a pro-innovation approach - white paper, mar

Innovation Department for Science and Technology. Ai regulation: a pro-innovation approach - white paper, mar

work page

[5] [5]

Digital empires: The global battle to regulate technology

Anu Bradford. Digital empires: The global battle to regulate technology. Oxford University Press, 2023

work page 2023

[6] [6]

From a ‘race to ai’to a ‘race to ai regulation’: regulatory competition for artificial intelligence

Nathalie A Smuha. From a ‘race to ai’to a ‘race to ai regulation’: regulatory competition for artificial intelligence. Law, Innovation and Technology, 13(1):57–84, 2021

work page 2021

[7] [7]

Artificial intelligence between national strategies and the global regulatory race: analytical routes for an international and comparative reappraisal

Pasquot Polido and Fabricio Bertini. Artificial intelligence between national strategies and the global regulatory race: analytical routes for an international and comparative reappraisal. Rev. Faculdade Direito Universidade Federal Minas Gerais, 76:229, 2020

work page 2020

[8] [8]

The global institutional governance of ai: A four -dimensional perspective

Rostam J Neuwirth. The global institutional governance of ai: A four -dimensional perspective. International Journal of Digital Law and Governance, 1(1):113–153, 2024

work page 2024