Contesting the algorithm: advancing a right to challenge AI decisions under the GDPR for algorithmic fairness
Bibliographic record
Abstract
Purpose This study aims to challenge the adequacy of Article 22 of the General Data Protection Regulation (GDPR) in safeguarding individuals against harmful automated decisions. It argues that explainability alone is insufficient for algorithmic accountability and proposes a legally enforceable right to contest such decisions. Through comparative legal analysis, it reveals the shortcomings of current GDPR protections and advocates for an amended framework that empowers individuals with substantive rights and remedies. The goal is to enable individuals not only to understand but also to challenge, correct or overturn artificial intelligence (AI)-driven decisions that significantly affect their lives. Design/methodology/approach This study adopts a qualitative comparative case study approach, analyzing enforcement and contestability in eight jurisdictions: four under the GDPR (The Netherlands, UK, France and Germany) and four outside it (California, New York City and Canada – public and private sectors). Data sources include legal texts, academic literature, court rulings and policy documents. A structured analytical matrix was applied to assess algorithm type, sector, availability of contestability mechanisms and enforcement effectiveness. This desk-based comparative legal analysis triangulates secondary sources to identify regulatory gaps and formulate reform proposals for strengthening contestability rights in AI governance. Findings The analysis reveals that GDPR Article 22 is functionally weak due to vague language, broad exceptions and limited enforcement. In practice, individuals rarely access meaningful mechanisms to contest consequential AI-driven decisions. By contrast, non-GDPR jurisdictions such as California and Canada show more proactive governance through opt-out rights, bias audits and algorithmic impact assessments. This study finds that effective contestability requires not only individual rights but also institutional safeguards, including human-in-the-loop review, independent oversight and public accountability mechanisms. Transparency alone is insufficient – robust, enforceable procedural rights are essential to ensure fairness and protect affected individuals. Originality/value This paper offers a novel ethical and legal case for rethinking algorithmic fairness beyond explainability, introducing a structured proposal to amend GDPR Article 22. It moves the discourse from transparency to contestability, grounded in comparative case analysis across EU and non-EU jurisdictions. The work bridges theoretical critique and practical reform, offering actionable policy recommendations, including an explicit right to contest, standards for human review and regulatory oversight models. It contributes original insights into how algorithmic harms can be addressed through due process-based contestation rights, reinforcing autonomy, fairness and justice in AI governance.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.086 | 0.150 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.006 | 0.061 |
| Scholarly communication | 0.017 | 0.011 |
| Open science | 0.003 | 0.009 |
| Research integrity | 0.006 | 0.006 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".