INTERNATIONAL HUMAN RIGHTS LAW AS A FRAMEWORK FOR ALGORITHMIC ACCOUNTABILITY
Open accessThe DOI matches its Crossref record; 11 of the 16 references it cites were found in a registry or an index; 1 of the 85 sources its claims cite were checked for support.
Abstract
Abstract Existing approaches to ‘algorithmic accountability’, such as transparency, provide an important baseline, but are insufficient to address the (potential) harm to human rights caused by the use of algorithms in decision-making. In order to effectively address the impact on human rights, we argue that a framework that sets out a shared understanding and means of assessing harm; is capable of dealing with multiple actors and different forms of responsibility; and applies across the full algorithmic life cycle, from conception to deployment, is needed. While generally overlooked in debates on algorithmic accountability, in this article, we suggest that international human rights law already provides this framework. We apply this framework to illustrate the effect it has on the choices to employ algorithms in decision-making in the first place and the safeguards required. While our analysis indicates that in some circumstances, the use of algorithms may be restricted, we argue that these findings are not ‘anti-innovation’ but rather appropriate checks and balances to ensure that algorithms contribute to society, while safeguarding against risks.
Overview
CoChat Summary
TL;DR
The article argues that existing algorithmic accountability approaches such as transparency are necessary but insufficient, and that international human rights law (IHRL) provides a shared framework for defining harm, allocating responsibilities, and covering the full algorithmic life cycle. Applying IHRL yields red lines, required safeguards, and heightened rather than reduced responsibility for unpredictable algorithmic effects.
Key findings
- Transparency, explainability, and technical fixes are a necessary baseline but incomplete; the authors identify gaps around defining harm, addressing the full algorithmic life cycle, allocating State/business responsibilities, providing remedies, and moving beyond after-the-fact accountability.
- IHRL supplies a framework: substantive and procedural rights define harm; States have obligations to prevent, protect, monitor, oversee, and remedy; businesses have expectations under the UN Guiding Principles on Business and Human Rights; and the framework applies from conception and design through deployment.
- Applying IHRL produces three effects: it may prohibit algorithms in certain decisions, such as where use would circumvent IHRL or where sentencing, bail, or parole decisions are made exclusively by algorithm due to arbitrariness and group-based correlation; it may require modifications or safeguards and thus delay deployment; and it rejects reduced responsibility where algorithmic effects cannot be predicted, potentially increasing responsibility.
- A human “in the loop” is not automatically an effective safeguard; the article’s discussion of *State of Wisconsin v Eric L. Loomis* and COMPAS illustrates risks of deference and inability to challenge proprietary algorithmic reasoning, so safeguards must address how the algorithm functions, not only input data and outcomes.
Methods
- Design: conceptual and doctrinal legal analysis; no empirical sample or dataset is reported.
- Data/sources: IHRL instruments and commentary, including UN Human Rights Committee General Comments, the Ruggie Principles, ECtHR cases, and existing algorithmic accountability literature.
- Main analysis: maps IHRL obligations and business expectations onto the algorithmic life cycle and applies the framework to illustrative contexts including criminal justice risk assessment, social security, child safeguarding, credit scoring, content moderation, and facial-image sexual-orientation inference.
Limitations
- The paper states that IHRL is not an exclusive or ready-made solution; business human rights responsibilities are still developing, IHRL does not currently impose direct obligations on businesses under international law, global business operations create regulatory and enforcement gaps, and IHRL cannot resolve all algorithmic accountability challenges, some of which belong to fields such as data protection.
- The paper states that its scope excludes discussion of established human rights mechanisms and that much work remains to operationalize and apply the IHRL framework.
- Evident from the text: the framework is not empirically tested or validated, and the examples are illustrative rather than a systematic sample; no quantitative outcomes are reported.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
1 Valid · 14 Not valid · 72 Not checked · 182 Own results and interpretations
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“As a result of a narrow conceptualization of harm, impact assessments in an algorithmic context have typically focused on issues relating to discrimination and privacy.”
What the sources say
- Not matchedcites K Crawford and J Schultz, 'Big Data and Due Process: Toward a Framework to Redress Predictive Privacy Harms' (2014) 55(1) Boston College Law Review 93, 95
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites Kroll et al. (n 9) 678.
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plainstates a cause- 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.
- Not matched
None of the sources cited for this claim could be matched to a paper in Crossref, so there is nothing to check it against.
In the paper
“This is important as States are increasingly reported to be integrating algorithms within their decisionmaking processes across a range of sectors that may have signi fi cant consequences for individuals and groups in areas such as policing, sentencing, social security and the identi fi cation of children at risk, as noted above.”
What the sources say
- Not matchedcites Canadian Institute for Advanced Research, 'CIFAR Pan-Canadian Artificial Intelligence Strategy'
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites E Macron, 'Artificial Intelligence: Making France a Leader'
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites NITI Aayog, 'National Strategy for Artificial Intelligence #AIFORALL'
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
hedged- 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.
- Not matched
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“While generally overlooked in debates on algorithmic accountability, in this article, we suggest that international human rights law already provides this framework.”
hedgedstates a cause- 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: unclear — It is unclear why the paper says it holds.
A statement the paper makes itself — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Yet, big datadriven algorithmic models may be used to make individually-focused decisions.”
hedgedstates a possible cause- 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.
In the paper
“However, the nature of big data-driven algorithms means that they generate results that describe group behaviour, but which are not tailored to speci fi c individuals within that group, irrespective of the size or quality of the input dataset. 30”
What the sources say
- Not matchedcites [30]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plainstates a cause- 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.
- Not matched
85 more claims
Create a free account to see all 269 claims and every source verdict.
- Probably not a substantial claim. 136
- It may state two separate claims as one. 84
Citations
Cites
Ananny, M., & Crawford, K. (2018) · New Media & Society · doi:10.1177/1461444816676645 · Cited by 1,836
On the Morality of Artificial Agents
Floridi, L., & Sanders, J. W. (2004) · Minds and Machines · doi:10.1023/b:mind.0000035461.63578.9d · Cited by 1,218
The responsibility gap: Ascribing responsibility for the actions of learning automata
Matthias, A. (2004) · Ethics and Information Technology · doi:10.1007/s10676-004-3422-1 · Cited by 1,101
Counterfactual explanations without opening the black box: automated decisions and the GDPR
Wachter, S., Mittelstadt, B., & Russell, C. (2017) · SSRN Electronic Journal · doi:10.2139/ssrn.3063289 · Cited by 1,061
Wang, Y., & Kosinski, M. (2018) · Journal of Personality and Social Psychology · doi:10.1037/pspa0000098 · Cited by 766
Algorithmic Accountability: Journalistic investigation of computational power structures
Diakopoulos, N. (2015) · Digital Journalism · doi:10.1080/21670811.2014.976411 · Cited by 733
(2014) · Digital Commons at University of Maryland Carey Law (University of Maryland Francis King Carey School of Law) · Cited by 524
Zarsky, T. (2016) · Science, Technology, & Human Values · doi:10.1177/0162243915605575 · Cited by 422
Kroll, J. A., Huey, J., Barocas, S., Felten, E. W., Reidenberg, J. R., Robinson, D. G., & Yu, H. (2017) · FLASH - Fordham Law Archive of Scholarship & History (Fordham University) · Cited by 299
Dodig Crnkovic, G., & Çürüklü, B. (2012) · Ethics and Information Technology · doi:10.1007/s10676-011-9278-2 · Cited by 90
Cited by
AI for humanitarian action: human rights and ethics
Pizzi, M., Romanoff, M., & Engelhardt, T. (2020) · International Review of the Red Cross · doi:10.1017/s1816383121000011 · Cited by 145
International migration management in the age of artificial intelligence
Beduschi, A. (2021) · Migration Studies · doi:10.1093/migration/mnaa003 · Cited by 141
“Identifying such gaps and acting on them would allow state authorities to prepare and adapt their reception conditions, thus complying with their legal obligations under international human rights law (IHRL).”
BackgroundInfluentialFrom human resources to human rights: Impact assessments for hiring algorithms
Yam, J., & Skorburg, J. A. (2021) · Ethics and Information Technology · doi:10.1007/s10676-021-09599-7 · Cited by 169
Koshiyama, A., Kazim, E., Treleaven, P., Rai, P., Szpruch, L., Pavey, G., Ahamat, G., Leutner, F., Goebel, R., Knight, A., et al. (2024) · Royal Society Open Science · doi:10.1098/rsos.230859 · Cited by 110
“As such, we believe this will stifle innovation.”
Beyond a Human Rights-Based Approach to AI Governance: Promise, Pitfalls, Plea
Smuha, N. A. (2020) · Philosophy & Technology · doi:10.1007/s13347-020-00403-w · Cited by 116
Metadata
- Authors
- McGregor, L., Murray, D., & Ng, V.
- Affiliations
- University of Essex (GB)
- Corresponding author
- Ng, Vivian
- Venue
- International and Comparative Law Quarterly
- Publisher
- Cambridge University Press (CUP)
- Published
- 2019-04
- Volume
- 68
- Issue
- 2
- Pages
- 309-343
- ISSN
- 0020-5893 (ISSN-L)
- 1471-6895
- OpenAlex
- W2940105172
- Cited by
- 207
- References
- 16
- Citations by year
- 2026: 12 · 2025: 36 · 2024: 26 · 2023: 72 · 2022: 36 · 2021: 14 · 2020: 9 · 2019: 2
- FWCI
- 26.66
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- International and Comparative Law Quarterly (publishedVersion, cc-by-nc-sa)
- Open Access at Essex (University of Essex) (acceptedVersion)
- Funders
- Economic and Social Research Council (ES/M010236/1)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Ethics and Social Impacts of AI
- Social Sciences › Law › Digitalization, Law, and Regulation
- Social Sciences › Law › Artificial Intelligence in Law
- Keywords
- algorithmic accountability, international human rights law, algorithmic decision-making, algorithmic transparency, harm assessment, algorithmic discrimination, algorithmic governance, algorithmic innovation
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.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
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: INTERNATIONAL HUMAN RIGHTS LAW AS A FRAMEWORK FOR ALGORITHMIC ACCOUNTABILITY. https://app.cochat.ai/papers/10.1017/s0020589319000046
- The paper
- McGregor, L., Murray, D., & Ng, V. (2019). INTERNATIONAL HUMAN RIGHTS LAW AS A FRAMEWORK FOR ALGORITHMIC ACCOUNTABILITY. International and Comparative Law Quarterly. https://doi.org/10.1017/s0020589319000046
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.