{"$schema":"https://policywindow.org/critique/api/schema","generated_by":"agi-social-scientist","url":"https://policywindow.org/critique/r/data-transparency","attestation":{"ok":true,"checks":[{"id":"coverage","label":"Coverage matrix re-derived from codings matches the published aggregate","pass":true,"detail":"22 codings → 5×6 matrix"},{"id":"gaps","label":"Re-derived empty cells are exactly the published gaps (identity, not just count)","pass":true,"detail":"18 re-derived; same cells; 18 published"},{"id":"included","label":"Unique included papers match the published inclusion count","pass":true,"detail":"22 re-derived vs 22 published"},{"id":"excluded","label":"Excluded records match the published exclusion count","pass":true,"detail":"12 re-derived vs 12 published"},{"id":"spans","label":"Every inclusion carries a verbatim rationale span (AGISS P1: ≥20 chars)","pass":true,"detail":"shortest span 70 chars (min 20)"}],"derived":{"coverage":{"algorithmic_decision_transparency":{"conceptual_normative":2,"legal_doctrinal":0,"empirical_qualitative":0,"empirical_quantitative":0,"technical_review_survey":2,"review_synthesis":1},"legal_regulatory_instrument":{"conceptual_normative":1,"legal_doctrinal":3,"empirical_qualitative":0,"empirical_quantitative":0,"technical_review_survey":0,"review_synthesis":0},"accountability_governance_mechanism":{"conceptual_normative":4,"legal_doctrinal":0,"empirical_qualitative":1,"empirical_quantitative":0,"technical_review_survey":0,"review_synthesis":2},"ethics_principles_frameworks":{"conceptual_normative":1,"legal_doctrinal":0,"empirical_qualitative":0,"empirical_quantitative":0,"technical_review_survey":1,"review_synthesis":3},"scientific_reproducibility_transparency":{"conceptual_normative":1,"legal_doctrinal":0,"empirical_qualitative":0,"empirical_quantitative":0,"technical_review_survey":0,"review_synthesis":0}},"totalCodings":22,"uniqueIncluded":22,"emptyCells":18,"excluded":12,"minSpanLength":70},"reportHash":"0f7198d6570a4d06","corpusHash":"010d24c0c28a579d"},"id":"data-transparency","reviewId":"CR-REV-005","agissStem":"data-transparency","question":"What does the scholarly literature report on data and algorithmic transparency in AI governance?","title":"Evidence gap map: data and algorithmic transparency in the AI-governance literature","reviewKind":"evidence_gap_map","method":"scoping_review_v1","headline":"Across a 5x6 framework matrix, 18 of 30 cells have no papers in this corpus of 22 included records; the populated cells concentrate in aspect 'accountability_governance_mechanism' (7 papers) and evidence type 'conceptual_normative' (9 papers).","verdict":"evidence_gap_map conducted over 34 records (22 included, 12 excluded with reasons): a coverage map with 18 named empty cells. Counts only — no importance adjudication (Sacred Rule 9); the report re-derives offline from corpus + codings.","coderModelId":"in-session:claude-agent","codingFrame":{"aspects":["algorithmic_decision_transparency","legal_regulatory_instrument","accountability_governance_mechanism","ethics_principles_frameworks","scientific_reproducibility_transparency"],"evidenceTypes":["conceptual_normative","legal_doctrinal","empirical_qualitative","empirical_quantitative","technical_review_survey","review_synthesis"]},"coverage":{"accountability_governance_mechanism":{"conceptual_normative":4,"empirical_qualitative":1,"empirical_quantitative":0,"legal_doctrinal":0,"review_synthesis":2,"technical_review_survey":0},"algorithmic_decision_transparency":{"conceptual_normative":2,"empirical_qualitative":0,"empirical_quantitative":0,"legal_doctrinal":0,"review_synthesis":1,"technical_review_survey":2},"ethics_principles_frameworks":{"conceptual_normative":1,"empirical_qualitative":0,"empirical_quantitative":0,"legal_doctrinal":0,"review_synthesis":3,"technical_review_survey":1},"legal_regulatory_instrument":{"conceptual_normative":1,"empirical_qualitative":0,"empirical_quantitative":0,"legal_doctrinal":3,"review_synthesis":0,"technical_review_survey":0},"scientific_reproducibility_transparency":{"conceptual_normative":1,"empirical_qualitative":0,"empirical_quantitative":0,"legal_doctrinal":0,"review_synthesis":0,"technical_review_survey":0}},"gaps":["algorithmic_decision_transparency x legal_doctrinal: 0 papers","algorithmic_decision_transparency x empirical_qualitative: 0 papers","algorithmic_decision_transparency x empirical_quantitative: 0 papers","legal_regulatory_instrument x empirical_qualitative: 0 papers","legal_regulatory_instrument x empirical_quantitative: 0 papers","legal_regulatory_instrument x technical_review_survey: 0 papers","legal_regulatory_instrument x review_synthesis: 0 papers","accountability_governance_mechanism x legal_doctrinal: 0 papers","accountability_governance_mechanism x empirical_quantitative: 0 papers","accountability_governance_mechanism x technical_review_survey: 0 papers","ethics_principles_frameworks x legal_doctrinal: 0 papers","ethics_principles_frameworks x empirical_qualitative: 0 papers","ethics_principles_frameworks x empirical_quantitative: 0 papers","scientific_reproducibility_transparency x legal_doctrinal: 0 papers","scientific_reproducibility_transparency x empirical_qualitative: 0 papers","scientific_reproducibility_transparency x empirical_quantitative: 0 papers","scientific_reproducibility_transparency x technical_review_survey: 0 papers","scientific_reproducibility_transparency x review_synthesis: 0 papers"],"disclosedFragilities":["scoping retrieval with fixed queries, not a systematic search (coverage is query-bounded)","coding from abstracts only — full texts were not consulted; cells count papers, not extracted effect estimates","single-annotator coding (in-session:claude-agent); no second coder, no kappa","corpus bounded to 34 retrieved records; counts are corpus-relative, not field-level claims"],"nCorpus":34,"nIncluded":22,"nExcluded":12,"reportHash":"0f7198d6570a4d06","corpusHash":"010d24c0c28a579d","selection":{"method":"review_selector_v1","selectedKind":"evidence_gap_map","features":{"corpus_kind":"mixed","full_texts_available":false,"has_quantitative_effects":false,"n_studies":34,"n_with_abstracts":30,"outcomes_comparable":false},"selectionHash":"0164a64dd68f6193","verdict":"Review-kind selection for 'What does the scholarly literature report on data and algorithmic transparency in AI governance?': SELECTED evidence_gap_map; 0 kind(s) rejected with their failed requirements recorded. A methodological screen (Sacred Rule 9): the selection is disclosed on the article and the selected kind's own discipline still applies at conduct time."},"includedCodings":[{"openalexId":"W2619904157","title":"Algorithmic Accountability and Public Reason","venue":"Philosophy & Technology","year":2017,"doi":"10.1007/s13347-017-0263-5","aspect":"accountability_governance_mechanism","evidenceType":"conceptual_normative","rationaleSpan":"I present an account of algorithmic accountability in terms of the democratic ideal of 'public reason'"},{"openalexId":"W2767837247","title":"Algorithmic Decision-Making Based on Machine Learning from Big Data: Can Transparency Restore Accountability?","venue":"Philosophy & Technology","year":2017,"doi":"10.1007/s13347-017-0293-z","aspect":"algorithmic_decision_transparency","evidenceType":"conceptual_normative","rationaleSpan":"Would transparency contribute to restoring accountability for such systems as is often maintained?"},{"openalexId":"W2786242872","title":"Fairness and Accountability Design Needs for Algorithmic Support in High-Stakes Public Sector Decision-Making","venue":"","year":2018,"doi":"10.1145/3173574.3174014","aspect":"accountability_governance_mechanism","evidenceType":"empirical_qualitative","rationaleSpan":"We interviewed 27 public sector machine learning practitioners across 5 OECD countries"},{"openalexId":"W2805992252","title":"Ethical Implications and Accountability of Algorithms","venue":"Journal of Business Ethics","year":2018,"doi":"10.1007/s10551-018-3921-3","aspect":"accountability_governance_mechanism","evidenceType":"conceptual_normative","rationaleSpan":"This article identifies whether developers have a responsibility for their algorithms later in use"},{"openalexId":"W2886443472","title":"Algorithms: transparency and accountability","venue":"Philosophical Transactions of the Royal Society A Mathematical Physical and Engineering Sciences","year":2018,"doi":"10.1098/rsta.2017.0351","aspect":"legal_regulatory_instrument","evidenceType":"conceptual_normative","rationaleSpan":"it looks at the legal protections for individuals afforded by the EU General Data Protection Regulation"},{"openalexId":"W2902634493","title":"AI4People—An Ethical Framework for a Good AI Society: Opportunities, Risks, Principles, and Recommendations","venue":"Minds and Machines","year":2018,"doi":"10.1007/s11023-018-9482-5","aspect":"ethics_principles_frameworks","evidenceType":"conceptual_normative","rationaleSpan":"present a synthesis of five ethical principles that should undergird its development and adoption"},{"openalexId":"W2954197146","title":"Transparency you can trust: Transparency requirements for artificial intelligence between legal norms and contextual concerns","venue":"Big Data & Society","year":2019,"doi":"10.1177/2053951719860542","aspect":"legal_regulatory_instrument","evidenceType":"legal_doctrinal","rationaleSpan":"We first investigate the ratio legis of the transparency requirement in the General Data Protection Regulation"},{"openalexId":"W2954503794","title":"DARPA's Explainable Artificial Intelligence Program","venue":"AI Magazine","year":2019,"doi":"10.1609/aimag.v40i2.2850","aspect":"algorithmic_decision_transparency","evidenceType":"technical_review_survey","rationaleSpan":"DARPA's explainable artificial intelligence (XAI) program endeavors to create AI systems whose learned models and decisions can be understood and appropriately trusted by end users"},{"openalexId":"W2958089299","title":"A Survey on Explainable Artificial Intelligence (XAI): Toward Medical XAI","venue":"IEEE Transactions on Neural Networks and Learning Systems","year":2020,"doi":"10.1109/tnnls.2020.3027314","aspect":"algorithmic_decision_transparency","evidenceType":"technical_review_survey","rationaleSpan":"Some of them require high level of accountability and thus transparency, for example, the medical sector"},{"openalexId":"W2989512989","title":"A governance model for the application of AI in health care","venue":"Journal of the American Medical Informatics Association","year":2019,"doi":"10.1093/jamia/ocz192","aspect":"accountability_governance_mechanism","evidenceType":"conceptual_normative","rationaleSpan":"concern has been expressed about the ethical and regulatory aspects of the application of AI in health care"},{"openalexId":"W3000603264","title":"The role of artificial intelligence in achieving the Sustainable Development Goals","venue":"Nature Communications","year":2020,"doi":"10.1038/s41467-019-14108-y","aspect":"accountability_governance_mechanism","evidenceType":"review_synthesis","rationaleSpan":"The fast development of AI needs to be supported by the necessary regulatory insight and oversight"},{"openalexId":"W3001807593","title":"Closing the AI accountability gap","venue":"","year":2020,"doi":"10.1145/3351095.3372873","aspect":"accountability_governance_mechanism","evidenceType":"conceptual_normative","rationaleSpan":"deployed systems are audited for harm by investigators from outside the organizations deploying the algorithms"},{"openalexId":"W3003385079","title":"The Ethics of AI Ethics: An Evaluation of Guidelines","venue":"Minds and Machines","year":2020,"doi":"10.1007/s11023-020-09517-8","aspect":"ethics_principles_frameworks","evidenceType":"review_synthesis","rationaleSpan":"this paper analyzes and compares 22 guidelines, highlighting overlaps but also omissions"},{"openalexId":"W3008058947","title":"Transparency and reproducibility in artificial intelligence","venue":"Nature","year":2020,"doi":"10.1038/s41586-020-2766-y","aspect":"scientific_reproducibility_transparency","evidenceType":"conceptual_normative","rationaleSpan":"we identify obstacles hindering transparent and reproducible AI research"},{"openalexId":"W3103319283","title":"Towards Transparency by Design for Artificial Intelligence","venue":"Science and Engineering Ethics","year":2020,"doi":"10.1007/s11948-020-00276-4","aspect":"algorithmic_decision_transparency","evidenceType":"conceptual_normative","rationaleSpan":"we develop the concept of Transparency by Design that serves as practical guidance in helping promote the beneficial functions of transparency"},{"openalexId":"W3124443940","title":"Why a Right to Explanation of Automated Decision-Making Does Not Exist in the General Data Protection Regulation","venue":"International Data Privacy Law","year":2017,"doi":"10.1093/idpl/ipx005","aspect":"legal_regulatory_instrument","evidenceType":"legal_doctrinal","rationaleSpan":"there are several reasons to doubt both the legal existence and the feasibility of such a right"},{"openalexId":"W3202183072","title":"Trustworthy AI: From Principles to Practices","venue":"ACM Computing Surveys","year":2022,"doi":"10.1145/3555803","aspect":"ethics_principles_frameworks","evidenceType":"technical_review_survey","rationaleSpan":"we provide AI practitioners with a comprehensive guide for building trustworthy AI systems"},{"openalexId":"W4300071243","title":"Slave to the Algorithm? Why a 'right to an explanation' is probably not the remedy you are looking for","venue":"","year":2017,"doi":"10.31228/osf.io/97upg","aspect":"legal_regulatory_instrument","evidenceType":"legal_doctrinal","rationaleSpan":"have caused a range of concerns revolving mainly around unfairness, discrimination and opacity"},{"openalexId":"W4315880904","title":"Ethics and governance of trustworthy medical artificial intelligence","venue":"BMC Medical Informatics and Decision Making","year":2023,"doi":"10.1186/s12911-023-02103-9","aspect":"ethics_principles_frameworks","evidenceType":"review_synthesis","rationaleSpan":"We adopted a multidisciplinary approach and summarized five subjects that influence the trustworthiness of medical AI"},{"openalexId":"W4381848566","title":"Connecting the dots in trustworthy Artificial Intelligence: From AI principles, ethics, and key requirements to responsible AI systems and regulation","venue":"Information Fusion","year":2023,"doi":"10.1016/j.inffus.2023.101896","aspect":"ethics_principles_frameworks","evidenceType":"review_synthesis","rationaleSpan":"Trustworthy Artificial Intelligence (AI) is based on seven technical requirements sustained over three main pillars"},{"openalexId":"W4400303336","title":"Transparency and accountability in AI systems: safeguarding wellbeing in the age of algorithmic decision-making","venue":"Frontiers in Human Dynamics","year":2024,"doi":"10.3389/fhumd.2024.1421273","aspect":"algorithmic_decision_transparency","evidenceType":"review_synthesis","rationaleSpan":"This review aims to provide an overview of the key legal and ethical challenges associated with implementing transparency and accountability in AI systems"},{"openalexId":"W4407857156","title":"Algorithmic bias, data ethics, and governance: Ensuring fairness, transparency and compliance in AI-powered business analytics applications","venue":"World Journal of Advanced Research and Reviews","year":2025,"doi":"10.30574/wjarr.2025.25.2.0571","aspect":"accountability_governance_mechanism","evidenceType":"review_synthesis","rationaleSpan":"concerns about fairness, transparency, and compliance have intensified"}],"excludedPapers":[{"openalexId":"W2582721310","title":"Why a Right to Explanation of Automated Decision-Making Does Not Exist in the General Data Protection Regulation","venue":"SSRN Electronic Journal","year":2016,"doi":"10.2139/ssrn.2903469","exclusionReason":"no abstract available for coding (abstracts-only protocol)"},{"openalexId":"W2770717476","title":"Exploring the impact of artificial intelligence on teaching and learning in higher education","venue":"Research and Practice in Technology Enhanced Learning","year":2017,"doi":"10.1186/s41039-017-0062-8","exclusionReason":"off-topic: AI in higher-education teaching, not transparency governance"},{"openalexId":"W2952426787","title":"Unraveling Transparency and Accountability in Blockchain","venue":"","year":2019,"doi":"10.1145/3325112.3325262","exclusionReason":"off-topic: blockchain for government transparency, not AI governance"},{"openalexId":"W2969625533","title":"Artificial Intelligence (AI): Multidisciplinary perspectives on emerging challenges, opportunities, and agenda for research, practice and policy","venue":"International Journal of Information Management","year":2019,"doi":"10.1016/j.ijinfomgt.2019.08.002","exclusionReason":"off-topic: broad AI overview; transparency governance is not its subject"},{"openalexId":"W2981731882","title":"Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI","venue":"Information Fusion","year":2019,"doi":"10.1016/j.inffus.2019.12.012","exclusionReason":"no abstract available for coding (abstracts-only protocol)"},{"openalexId":"W2981863007","title":"Systematic review of research on artificial intelligence applications in higher education – where are the educators?","venue":"International Journal of Educational Technology in Higher Education","year":2019,"doi":"10.1186/s41239-019-0171-0","exclusionReason":"off-topic: systematic review of AI in higher education"},{"openalexId":"W2982580298","title":"Key challenges for delivering clinical impact with artificial intelligence","venue":"BMC Medicine","year":2019,"doi":"10.1186/s12916-019-1426-2","exclusionReason":"off-topic: clinical-translation challenges of healthcare AI; transparency is not the abstract's subject"},{"openalexId":"W3017357696","title":"Transparency and accountability in AI decision support: Explaining and visualizing convolutional neural networks for text information","venue":"Decision Support Systems","year":2020,"doi":"10.1016/j.dss.2020.113302","exclusionReason":"no abstract available for coding (abstracts-only protocol)"},{"openalexId":"W3036911563","title":"Data governance: Organizing data for trustworthy Artificial Intelligence","venue":"Government Information Quarterly","year":2020,"doi":"10.1016/j.giq.2020.101493","exclusionReason":"no abstract available for coding (abstracts-only protocol)"},{"openalexId":"W3135539146","title":"Sustainable AI: AI for sustainability and the sustainability of AI","venue":"AI and Ethics","year":2021,"doi":"10.1007/s43681-021-00043-6","exclusionReason":"off-topic: sustainability of the AI lifecycle, not transparency governance"},{"openalexId":"W4360620450","title":"Opinion Paper: “So what if ChatGPT wrote it?” Multidisciplinary perspectives on opportunities, challenges and implications of generative conversational AI for research, practice and policy","venue":"International Journal of Information Management","year":2023,"doi":"10.1016/j.ijinfomgt.2023.102642","exclusionReason":"off-topic: broad ChatGPT opportunities/challenges collection"},{"openalexId":"W4360836968","title":"Sparks of Artificial General Intelligence: Early experiments with GPT-4","venue":"arXiv (Cornell University)","year":2023,"doi":"10.48550/arxiv.2303.12712","exclusionReason":"off-topic: capability evaluation of GPT-4, not transparency governance"}],"narrativeMarkdown":"# Evidence gap map: data and algorithmic transparency in the AI-governance literature\n\n## In plain terms\n\nWhere has scholarship on AI transparency actually looked — and where has it not? The engine retrieved a corpus on data and algorithmic transparency in AI governance, coded every abstract into a framework matrix, and counted. The picture: plenty of conceptual argument about accountability and governance mechanisms; very little measurement. More than half the matrix is empty.\n\n**Finding (screened coverage map):** Across a 5x6 framework matrix, 18 of 30 cells have no papers in this corpus of 22 included records; the populated cells concentrate in aspect 'accountability_governance_mechanism' (7 papers) and evidence type 'conceptual_normative' (9 papers).\n\n## Background\n\nPolicy Window's research engine selected the review kind BEFORE this article was drafted: a review-kind selector matched the question (\"What does the scholarly literature report on data and algorithmic transparency in AI governance?\") and the measured corpus features (34 records, 30 with abstracts, no comparable quantitative effects, abstracts only) to an evidence gap map; quantitative-synthesis kinds were rejected with their failed requirements recorded (selection `0164a64dd68f6193`). The article's status is a screened candidate routed to human review, not an adjudicated truth (Sacred Rule 9).\n\n## Method\n\nEvery record was coded exactly once against a declared frame — five transparency aspects by six evidence types — with a verbatim rationale span per inclusion and a stated reason per exclusion (12 records excluded: off-topic or no abstract). The coverage table is arithmetic over those codings. Review report `0f7198d6570a4d06`; corpus `010d24c0c28a579d`.\n\n## What the map shows\n\nThe populated region: conceptual and normative work on accountability and governance mechanisms, legal-doctrinal analysis of regulatory instruments (the GDPR right-to-explanation debate), technical surveys of explainable-AI methods, and review work on ethics principles. The named empty cells include every transparency aspect crossed with empirical_quantitative evidence — in this corpus, no included paper reports quantitative measurement of transparency outcomes. The one empirical study included is interview-based.\n\n## Limitations (disclosed by the engine)\n\n- scoping retrieval with fixed queries, not a systematic search (coverage is query-bounded)\n- coding from abstracts only — full texts were not consulted; cells count papers, not extracted effect estimates\n- single-annotator coding (in-session:claude-agent); no second coder, no kappa\n- corpus bounded to 34 retrieved records; counts are corpus-relative, not field-level claims\n\n## Verify\n\nEvery count above re-derives offline from the committed corpus and codings — no model, no network, no trust in the institute required: `PYTHONPATH=src python scripts/verify_review.py  # exit 0 = re-derives offline`\n","verifyCommand":"PYTHONPATH=src python scripts/verify_review.py data-transparency  # exit 0 = re-derives offline","aiAgiCategories":["AI_governance","knowledge_production"],"publicationDate":"2026-06-14","published":true}