Artificial intelligence and human rights: Between law and ethics

Analysis

Why This Paper Matters

This paper addresses a critical gap in AI governance: the often-fragmented relationship between ethical guidelines and legal regulations. As AI systems become more pervasive, practitioners and policymakers face the challenge of reconciling voluntary ethical commitments with binding legal requirements. Sartor's work provides a conceptual bridge, arguing that human rights and social values can serve as a common normative foundation. This is particularly relevant for AI practitioners who must navigate both ethical and legal constraints in their work.

The paper's significance lies in its systematic examination of how law and ethics can both overlap and diverge. By highlighting the dialectical tension between the two, Sartor moves beyond simplistic calls for 'ethical AI' or 'legal compliance' and instead offers a nuanced perspective that acknowledges the dynamic interplay between normative systems. This is essential for developing robust governance frameworks that are both principled and practical.

Technical Contributions

• Normative Framework: Proposes human/fundamental rights as a shared vocabulary for ethics and law, enabling better coordination.
• Dialectical Tension: Introduces the concept of productive tension between legal and ethical demands, rather than assuming they must always align.
• Socio-Technical Perspective: Emphasizes that AI systems are embedded in social and technical contexts, requiring governance that addresses both.
• Comparative Analysis: Systematically compares how ethical principles and legal rules address similar AI issues, identifying areas of convergence and conflict.

Results

As a conceptual paper, the 'results' are primarily theoretical. The main outcome is a well-argued thesis that human rights can serve as a unifying normative reference. The paper does not present empirical data or quantitative metrics, but it offers a clear analytical framework that can be used by researchers and policymakers to evaluate AI governance proposals. The argument is supported by references to legal and ethical theory, though specific case studies are not provided.

Significance

The paper contributes to the growing literature on AI ethics and governance by providing a conceptual foundation for aligning ethical and legal approaches. For AI practitioners, it underscores the importance of considering both ethical principles and legal obligations in system design and deployment. The framework could inform the development of more coherent regulatory policies, helping to avoid conflicts between voluntary ethical codes and binding laws. Moreover, by emphasizing human rights, the paper aligns AI governance with broader societal values, which is crucial for maintaining public trust. This work is likely to be influential in academic and policy circles, as it offers a way to harmonize the often-disparate discussions around AI ethics and law.