Toward Trustworthy Artificial Intelligence in Law and Public Administration: Transparency, Accountability and Decision Support
Open accessThe DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index; 5 of the 11 sources its claims cite were checked for support.
Abstract
This study investigates how trustworthy artificial intelligence can be operationalised in law and public administration through the interlocking principles of transparency, accountability, and decision support. The central argument is that public-sector AI cannot be treated as a purely technical efficiency instrument. In legal and administrative settings, algorithmic systems affect due process, reasons-giving, procedural fairness, access to services, and the legitimacy of state action. Accordingly, trustworthy deployment depends not only on computational capability but also on the institutional environment within which AI is designed, procured, documented, reviewed, and contested. The analysis therefore integrates doctrinal and policy review with a real-data cross-jurisdictional comparison covering Denmark, Estonia, the United Kingdom, the United States, Singapore, Brazil, India, and South Africa. The empirical component combines three publicly available indicators: the Oxford Insights Government AI Readiness Index 2024, the United Nations E-Government Development Index 2024, and the World Justice Project Rule of Law Index 2024. These real-data sources are used to construct a composite Trustworthy Public AI Index and to estimate exploratory association patterns between AI readiness, digital-government maturity, and rule-of-law performance. The results show that jurisdictions with stronger rule-of-law and digital-government foundations generally exhibit stronger conditions for trustworthy public-sector AI, although high AI readiness alone does not guarantee a proportionate level of legal accountability. The findings support a governance model in which documentation, human oversight, disclosure standards, reviewability, and institutional audit capacity are treated as core design requirements rather than after-the-fact safeguards. Keywords trustworthy artificial intelligence; law; public administration; transparency; accountability; decision support; rule of law; digital government; algorithmic governance; explainability
Overview
CoChat Summary
TL;DR
This paper argues that trustworthy AI in law and public administration requires transparency, accountability, and meaningful decision support embedded in legal-institutional and digital-government capacity, not technical readiness alone. Using 2024 cross-jurisdictional indicators for eight jurisdictions, it finds that stronger rule-of-law and digital-government foundations are associated with stronger conditions for trustworthy public-sector AI.
Key findings
- The composite Trustworthy Public AI Index ranks Denmark first at 86.00, followed by Singapore at 85.23; the United Kingdom, United States, and Estonia form an upper-middle cluster between 82.07 and 82.79, while Brazil, South Africa, and India range from 59.32 to 64.86.
- Correlations are positive across variables: rule of law correlates with the composite trust index at 0.939, digital-government maturity with the composite at 0.886, AI readiness with rule of law at 0.640, and AI readiness with digital-government maturity at 0.591.
- The exploratory OLS model estimates AI readiness as a function of rule of law and digital-government maturity; both coefficients are positive (rule of law 0.378, EGDI 0.162), with R-squared 0.416, adjusted R-squared 0.182, F-statistic 1.779, and Prob(F-statistic) 0.261.
- The paper concludes that high AI readiness alone does not guarantee proportionate legal accountability, and that documentation, human oversight, disclosure standards, reviewability, and audit capacity should be core design requirements.
Methods
- Mixed doctrinal-empirical design: doctrinal review of international governance instruments including the OECD AI Principles, NIST AI Risk Management Framework, UNESCO Recommendation, Council of Europe AI Convention, EU AI Act, UK Algorithmic Transparency Recording Standard, US OMB Memorandum M-2410, Singapore’s Model AI Governance Framework, Brazil’s Plano Brasileiro de Inteligência Artificial, and India’s AI for All policy lineage.
- Empirical component uses a purposive sample of eight jurisdictions — Denmark, Estonia, the United Kingdom, the United States, Singapore, Brazil, India, and South Africa — with 2024 data from the Oxford Insights Government AI Readiness Index, the United Nations E-Government Development Index, and the World Justice Project Rule of Law Index; UN and WJP scores are rescaled from 0–1 to 0–100.
- Analysis constructs a composite index weighted 40% AI readiness, 35% rule of law, and 25% digital-government maturity, then uses descriptive statistics, a correlation matrix, and an exploratory ordinary least squares regression.
Limitations
- Stated: cross-national indices capture broad institutional conditions rather than direct measures of every public AI system, so the data speak to enabling environments rather than case-level legality; index methodologies differ in construction, coverage, and update cycles, and any composite embeds judgment about weighting and comparability; the small purposive sample means the quantitative section is analytically illustrative and hypothesis-generating.
- Evident from the text: the OLS model has only eight observations, non-significant p-values (Prob(F-statistic) = 0.261), and a modest adjusted R-squared of 0.182, so the regression supports directional association rather than causal inference.
- Evident from the text: the composite index weights are author-defined, and the paper itself notes that measurement should remain subordinate to context-specific legal review, stakeholder engagement, and impact assessment.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
5 Valid · 3 Not valid · 4 Not checked · 86 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“This subtheme also has global and multi-sector relevance because the same governance tensions now appear in welfare administration, local government, insurance analytics, and health planning.”
What the sources say
- Partly supportedcites Rodriguez Müller, Tangi & Lerusse 2025 sourceExperimental — accuracy not yet measured
The Introduction discusses AI decision-making in local government and gives welfare-applicant assessments as an example; Section 2.2 identifies governance concerns including transparency, accountability and fairness. The paper does not establish that these tensions appear in insurance analytics or health planning. Its Conclusions also limit the study itself to Belgian local-government road maintenance. (LLM, conf=0.87)
- Not matchedcites Luna 2025
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
hedgedstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The results show that jurisdictions with stronger rule-of-law and digital-government foundations generally exhibit stronger conditions for trustworthy public-sector AI, although high AI readiness alone does not guarantee a proportionate level of legal accountability.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The results show that jurisdictions with stronger rule-of-law and digital-government foundations generally exhibit stronger conditions for trustworthy public-sector AI, although high AI readiness alone does not guarantee a proportionate level of legal accountability.”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The United States is a notable case of very high AI readiness combined with a rule-of-law score below the leading European cases, while Singapore combines high AI readiness with strong digital-government maturity and a solid, though not Denmark-level, rule-of-law position.”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Figure 4.2 shows that the sample broadly trends upward: jurisdictions with stronger rule-of-law scores tend to record higher AI readiness.”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
29 more claims
Create a free account to see all 98 claims and every source verdict.
- Probably not a substantial claim. 43
- It may state two separate claims as one. 27
- Borderline: one of the tests for a substantial claim is unclear. 2
Citations
Cites
The publisher hasn’t deposited this paper’s reference list.
Research works printed but not linked: 2
- Tomachуnskyi, SM, Kovaleva, NM & Adler, OG 2021, 'Artificial intelligence as an inventor: state, problems, development prospects', Expert: Paradigm of Law and Public Administration, vol. 17, no. 5, pp. 103-110.
- Pasupuleti, MK 2023c, 'Foundation Models in PyTorch: Retrieval-Augmented FineTuning with Safety Guardrail', International Journal of Academic and Industrial Research Innovations (IJAIRI), vol. 03, no. 03, pp. 1-16.
Cited by
Pasupuleti, M. K. (2026) · International Journal of Academic and Industrial Research Innovations(IJAIRI) · doi:10.62311/nesx/rp3sep-30092026
Self-citation
Metadata
- Authors
- Pasupuleti, M. K.
- Affiliations
- National Education Services (IN)
- Corresponding author
- Pasupuleti, Murali Krishna
- Venue
- International Journal of Academic and Industrial Research Innovations(IJAIRI)
- Publisher
- National Education Services
- Published
- 2026-03-30
- Volume
- 06
- Issue
- 03
- Pages
- 1-28
- ISSN
- 3049-2343 (ISSN-L)
- OpenAlex
- W7147106924
- Semantic Scholar
- 004746aebf934c94d1101bd2fc4e01697f0204fe
- Cited by
- 0
- References
- 0
- FWCI
- 0.00
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- public-domain
- Locations
- International Journal of Academic and Industrial Research Innovations(IJAIRI) (publishedVersion, public-domain)
- Topics
- Social Sciences › Political Science and International Relations › E-Government and Public Services
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Social Sciences › Law › Digitalization, Law, and Regulation
- Keywords
- trustworthy AI, public sector AI, transparency, accountability, decision support, rule of law, e-government, AI readiness, algorithmic governance, human oversight, explainability, due process, procedural justice, AI governance
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-06, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-06 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Toward Trustworthy Artificial Intelligence in Law and Public Administration: Transparency, Accountability and Decision Support. https://app.cochat.ai/papers/10.62311/nesx/rp-1-28-30032026
- The paper
- Pasupuleti, M. K. (2026). Toward Trustworthy Artificial Intelligence in Law and Public Administration: Transparency, Accountability and Decision Support. International Journal of Academic and Industrial Research Innovations(IJAIRI). https://doi.org/10.62311/nesx/rp-1-28-30032026
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.