Administrative Liability for Harm Arising from Intelligent Automated Administrative Decisions
Open accessThe DOI matches its Crossref record; 2 of the 2 references it cites were found in a registry or an index; 1 of the 24 sources its claims cite were checked for support.
Abstract
Administrative work is witnessing a radical transformation due to reliance on artificial intelligence techniques in decision making, which has given rise to a new category of decision-making: automated administrative decisions issued without direct human intervention. Despite the efficiency and speed, they provide, these decisions pose profound legal questions about their legitimacy, the guarantees that accompany them, and the scope of administrative liability for any resulting harm. This article examines the nature of this liability, its limits, and its existence in the absence of clear legal regulation in many legal systems. It also exposes a legal vision aimed at protecting individuals and reconciling technological progress with the principles of legitimacy and accountability.
Overview
CoChat Summary
TL;DR
This article argues that intelligent automated administrative decisions remain administrative acts subject to legality and accountability, so the administration cannot escape liability merely because AI or algorithms produced the decision. It examines fault-based, strict, and joint liability, and calls for specialised legislation, transparency, data protection, oversight, and human review.
Key findings
- Intelligent automated administrative decisions still require the classic elements of administrative acts—competence, form/procedure, cause, object, and purpose—and the administration remains responsible for selecting, operating, and monitoring the system.
- Liability may be fault-based (requiring fault, damage, and causal link), strict/non-fault (requiring only a causal link between the automated decision and damage), or joint/shared with designers, programmers, operators, or providers; the paper cites GDPR Article 82(4)-(5) as an analogous allocation model.
- Transparency, personal-data protection, and oversight are essential safeguards; the paper states that the GDPR mandates human intervention when an automated decision produces significant legal effects, and cites Article 24(1) GDPR on controller obligations.
- Many national regimes, particularly in the Arab world, have not kept pace, and the paper recommends specialised AI legislation plus post-decision human review for decisions affecting fundamental rights.
Methods
- Doctrinal legal study using descriptive and analytical approaches, presenting and examining relevant legal texts and juristic opinions.
- No empirical sample, dataset, or statistical analysis is reported; the analysis is organised around the legal nature of intelligent automated administrative decisions, their categories and safeguards, and the bases of administrative liability.
Limitations
- The paper states that the lack of a precise legal framework in many legal systems risks undermining individuals’ fundamental guarantees, especially in many Arab regimes.
- The paper states that establishing fault and causal link is difficult because of the “black-box” nature of certain algorithmic systems, whose internal logic and decision pathways are opaque.
- Evident from the text: the study offers normative legal analysis and recommendations without empirical data, case samples, or quantitative validation.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
1 Valid · 3 Not valid · 23 Not checked · 54 Own results and interpretations
In the paper
“This confirms the axiom that the administration remains fully liable for operating a faulty algorithm (19) .”
§ Administrative Liability for Harm Arising from Intelligent Automated Administrative Decisions > 2.1.1. Digital Technology Imposes New Administrative Obligations
What the sources say
- Not checked yetcites [19] 19. Mnsl, K., & Shāwish, Ḥ. (2021). Al-ishkālāt al-qānūnīyah li-iʿtimād al-qarār al-idārī al-khuwārizmī ["Legal issues of relying on algorithmic administrative decisions"]. El-Ryssala Journal for Studies and Researches in Humanities, 6(4), [page range unknown].
states 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.
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
“The final act is therefore attributed to the body that designed or adopted the algorithmic system, not to the technological tool itself.”
§ Administrative Liability for Harm Arising from Intelligent Automated Administrative Decisions > 1.1.3. Elements of the Intelligent Automated Administrative Decision
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.
In the paper
“If, however, the authority enjoys discretion, human intervention becomes possible (9) .”
§ Administrative Liability for Harm Arising from Intelligent Automated Administrative Decisions > 1.1.3. Elements of the Intelligent Automated Administrative Decision
What the sources say
- Not matchedcites [9] 9. Ṣā' im Aḥmad, R. M. (2022). Taṭbīqāt al-idārah lil-dhakāʾ al-iṣṭināʿī fī ittikhādh al-qarārāt al-idāriyyah (Master's dissertation). Middle East University.
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
hedgedstates a possible cause- 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.
- 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
“Accordingly, the purpose of an intelligent automated administrative decision is fixed at the stage when human designers program the processing system.”
§ Administrative Liability for Harm Arising from Intelligent Automated Administrative Decisions > 1.1.3. Elements of the Intelligent Automated Administrative Decision
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.
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
“This category of decisions is far harder to automate, because it usually demands adaptable reasoning and a nuanced appraisal of the circumstances, something even sophisticated systems struggle to deliver.”
§ Administrative Liability for Harm Arising from Intelligent Automated Administrative Decisions > 1.2.1. Categories of intelligent Automated Administrative Decisions
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.
20 more claims
Create a free account to see all 81 claims and to verify the 14 references still open.
- It may state two separate claims as one. 30
- Probably not a substantial claim. 42
- The quote only partly states this claim. 1
Citations
Cites
The publisher hasn’t deposited this paper’s reference list.
Research works printed but not linked: 18
- [2]2. -Muḥammad Ḥamdī ʻAlī ʻUmar, al-taʻwīḍ ʻan al-aḍrār fī majāl al-Masʼūlīyah al-Idārīyah, Journql of legal and economic research, Menoufia University, Volume55, N3, 2022, p.360.
- [3]3. -Muḥammad Balkhayr Ayt ʻwdyh, al-Masʼūlīyah ʻan aḍrār al-qarārāt al-Idārīyah al-Khuwārizmīyah, Academic Journal of Legal Research, Volume13, N1, 2022,p.279.
- [4]4. -Fahd Saʻīd alẓhwry, Muṣṭafá Sālim al-Nujayfī, Masʼūlīyat al-Idārah ʻan Istikhdāmāt al-dhakāʼ alāṣṭnāʻy ʻalá Asās al-khaṭaʼ, University of Sharjah's Journal of Law sciences, Volume21, N1, 2022, p.316.
- [6]6. -"The term 'black box' signifies a software system akin to a sealed, solid box whose inner workings cannot be grasped and whose behaviour cannot be explained." Yavar Bathaee, The Artificial Intelligence Black Box and the failure of intent and causation, Harvard journal of law and technology, volume 31, 2018.
- [8]8. Ḥamīd, K. (2021-2022). Taṭbīqāt al-dhakāʾ al-iṣṭināʿī fī nashāṭ al-idārah al-ʿāmah (Unpublished doctoral thesis). Faculty of Law and Political Sciences, University of Ghardaïa.
- [9]9. Ṣā' im Aḥmad, R. M. (2022). Taṭbīqāt al-idārah lil-dhakāʾ al-iṣṭināʿī fī ittikhādh al-qarārāt al-idāriyyah (Master's dissertation). Middle East University.
- [10]10. -Bilāl Aḥmad Salāmah Badr, Masʼūlīyat al-dawlah ʻan aḍrār al-dhakāʼ alāṣṭnāʻy, Journal of legal and economic sciences, Volume66, N3, Egypt, 2024,p.1429.
- [12]12. Al-Shammarī, A. ʿA. Z. Ḥ. (2024). Aḥkām al-qarār al-idārī al-iliktrūnī (dirāsah muqāranah) ["Provisions of the electronic administrative decision (comparative study)"]. Misan Journal for Comparative Legal Studies, 1(10), [page range unknown].
- [13]13. ʿAbbās, A. N. (2024). Al-qarārāt al-idāriyyah al-mūtamath, wa al-sulṭah al-taqdīrīyah li-jihāt al-idārah ["The administrative decisions executed, and the discretionary authority of administrative bodies"]. Journal of Legal and Economic Studies, 3, [page range unknown].
- [14]14. ʿAbd al-ʿAzīz, ʿA. Ṭ. (2010). Arkān al-qarār al-idārī al-iliktrūnī ["Elements of the electronic administrative decision"]. The Law Journal for Research and Studies, 2, [page range unknown].
- [16]16. Salāmah Badr, B. A. (2024). Masʾūlīyat al-dawlah ʿan aḍrār al-dhakāʾ al-iṣṭināʿī ["State liability for artificial intelligence harms"]. Journal of Legal and Economic Sciences, 66(3), [page range unknown].
- [18]18. Alẓhwry, F. S., & al-Nujayfī, M. S. (2022). Masʾūlīyat al-idārah ʿan istikhḍāmāt al-dhakāʾ al-iṣṭināʿī ʿalā asās al-khaṭaʾ ["Liability of administration for AI uses on the basis of error"]. University of Sharjah's Journal of Law Sciences, 21(1), [page range unknown].
- [19]19. Mnsl, K., & Shāwish, Ḥ. (2021). Al-ishkālāt al-qānūnīyah li-iʿtimād al-qarār al-idārī al-khuwārizmī ["Legal issues of relying on algorithmic administrative decisions"]. El-Ryssala Journal for Studies and Researches in Humanities, 6(4), [page range unknown].
- [20]20. Ayt ʿWdyh, M. B. (2022). Al-masʾūlīyah ʿan aḍrār al-qarārāt al-idāriyyah al-khuwārizmīyah ["Liability for damage of algorithmic administrative decisions"]. Academic Journal of Legal Research, 13(1), [page range unknown].
- [21]21. ʿUmar, M. Ḥ. A. (2022). Al-taʿwīḍ ʿan al-aḍrār fī majāl al-masʾūlīyah al-idāriyyah ["Compensation for damage in the field of administrative liability"]. Journal of Legal and Economic Research, 55(3), [page range unknown].
- [23]23. Maḥrūs, U. ʿA. (2024). Al-masʾūlīyah al-idāriyyah tujāha akhṭāʾ taṭbīqāt al-dhakāʾ al-iṣṭināʿī ["Administrative liability towards errors in applications of artificial intelligence"]. Journal of Law, 22(1), [page range unknown].
- [26]26. Sirrīyah, S. (2024). Taṭbīq tiqniyāt al-dhakāʾ al-iṣṭināʿī fī majāl al-qarārāt al-idāriyyah ["Application of artificial intelligence techniques in the field of administrative decisions"]. Al-Turath Journal, 14(2), [page range unknown].
- [27]27. Bathaeé, Y. (2018). The artificial intelligence black box and the failure of intent and causation. Harvard Journal of Law & Technology, 31(2), 890-938. https://doi.org/10.2139/ssrn.2857237 People's Democratic Republic of Algeria. (2018, June 10). Law No. 18-07 on the protection of natural persons in the processing of personal data. Official Journal of the People's Democratic Republic of Algeria, No. 34. Castets Renard, C. (2019). Encadrement des risques techniques et juridiques des activités de police prédictive ["Regulation of technical and legal risks of predictive policing activities"]. CHEMI, [volume/page unknown]. · doi:10.2139/ssrn.2857237
Metadata
- Authors
- Somia, K.
- Affiliations
- Université d'Oran 2 (DZ)
- Corresponding author
- Somia, KHADER
- Venue
- Science of Law
- Published
- 2025-10-25
- Volume
- 2025
- Issue
- 4
- Pages
- 71-77
- ISSN
- 2995-8717 (ISSN-L)
- OpenAlex
- W4415839578
- Cited by
- 0
- References
- 0
- FWCI
- 0.00
- Type
- journal article
- Language
- en
- Open access
- diamond
- doi.org
- Licence
- Locations
- Science of law. (publishedVersion, cc-by-nc-sa)
- Topics
- Social Sciences › Law › Digitalization, Law, and Regulation
- Social Sciences › Law › Administrative Law and Governance
- Social Sciences › Law › Artificial Intelligence in Law
- Keywords
- administrative liability, automated administrative decisions, artificial intelligence, algorithmic decision-making, accountability, AI regulation, personal protection
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 2026-09-29, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-09-29 (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: Administrative Liability for Harm Arising from Intelligent Automated Administrative Decisions. https://app.cochat.ai/papers/10.55284/w421rk10
- The paper
- Somia, K. (2025). Administrative Liability for Harm Arising from Intelligent Automated Administrative Decisions. Science of Law. https://doi.org/10.55284/w421rk10
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.