Rules, Cases, and Reasoning: Positivist Legal Theory as a Framework for Pluralistic AI Alignment
Not on the map. Why?
Abstract (arXiv)
Legal theory can address two related key problems of alignment: pluralism and specification. Alignment researchers must determine how to specify what is concretely meant by vague principles like helpfulness and fairness and they must ensure that their techniques do not exclude alternative perspectives on life and values. The law faces these same problems. Leading legal theories suggest the law solves these problems through the interaction of rules and cases, where general rules promulgated by a democratic authority are given specific content through their application over time. Concrete applications allow for convergence on practical meaning while preserving space for disagreement on values. These approaches suggest improvements to existing democratic alignment processes that use AI to create cases that give content to rules, allowing for more pluralist alignment.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 21 of its 49 references that are in the map. First posted on arXiv 7 Oct 2024.
Builds on
21 of 49 references.
References · 21
- 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
- 2022Constitutional AI: Harmlessness from AI Feedbackcited by 2,950
- 2021A General Language Assistant as a Laboratory for Alignmentcited by 1,023
- 2022Discovering Language Model Behaviors with Model-Written Evaluationscited by 805
- 2020Aligning AI With Shared Human Valuescited by 706
- 2020Artificial Intelligence, Values and Alignmentcited by 443
- 2018AI safety via debatecited by 384
- 2023Pretraining Language Models with Human Preferencescited by 245
- 2022Emergent Analogical Reasoning in Large Language Modelscited by 239
- 2022Measuring Progress on Scalable Oversight for Large Language Modelscited by 212
- 2023The Capacity for Moral Self-Correction in Large Language Modelscited by 171
- 2021Alignment of Language Agentscited by 167
9 more
- 2024The PRISM Alignment Dataset: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Modelscited by 136
- 2022Jury Learning: Integrating Dissenting Voices into Machine Learning Modelscited by 115
- 2023Value Kaleidoscope: Engaging AI with Pluralistic Human Values, Rights, and Dutiescited by 97
- 2023When the Majority is Wrong: Modeling Annotator Disagreement for Subjective Taskscited by 89
- 2023Personalisation within bounds: A risk taxonomy and policy framework for the alignment of large language models with personalised feedbackcited by 86
- 2022Training Language Models with Language Feedbackcited by 48
- 2023From Pseudorandomness to Multi-Group Fairness and Backcited by 13
- 2023Case Repositories: Towards Case-Based Reasoning for AI Alignmentcited by 11
- 2024Legal Minds, Algorithmic Decisions: How LLMs Apply Constitutional Principles in Complex Scenarioscited by 7
Built on by
Citing papers per year: 2025: 1; 2026: 1.
2 papers.
Similar, not cited
Close in wording, no citation either way
- 2026Alignment as Jurisprudence
- 2026Statutory AI: Aligning Large Language Models With Legal Norms
- 2024Dynamic Normativity: Necessary and Sufficient Conditions for Value Alignment
- 2026Towards a New Grammar of Reasoning for Artificial Legal Intelligence and the Mecelle as Its Semantic Protocol
- 2025The Law-Following AI Framework: Legal Foundations and Technical Constraints. Legal Analogues for AI Actorship and technical feasibility of Law Alignment
- 2026Position: Align AI to Our Aspirations, Not Our Flaws
- 2024Being Considerate as a Pathway Towards Pluralistic Alignment for Agentic AI
- 2026Pluralism in AI Governance: Toward Sociotechnical Alignment and Normative Coherence ↗