# Summary: 2026-07-27_13-06-38Z_RegulatingforAILegitimacy.md
Saved: 2026-07-28 00:11
Source: 2026-07-27_13-06-38Z_RegulatingforAILegitimacy.md
Model: None

---

## Summary  
The paper argues that AI legitimacy — the belief subjects have a rightful authority to be governed by AI systems — is an autonomous regulatory objective distinct from technical alignment and cannot be achieved merely through safety improvements. It identifies three failure points of AI legitimacy: opacity, private power, and administrative automation. The authors propose thin legality (publicity, stability) as a first step toward recognition, while thick legality supplies the missing public authorship of rules. They also outline three portable principles — integration, familiarity, and contestation — that can restore trust.

## Semantic links

## Key Contributions  
- Finding 1: Legitimacy is a sociological belief separate from alignment.  
- Finding 2: AI legitimacy falters at three sites: opacity, private power, and administrative automation.  
- Finding 3: Legal solutions include thin legality (publicity, stability) and thick legality (public authorship), plus the principles of integration, familiarity, and contestation.

## Methodology  
The authors approach the problem theoretically by analyzing existing AI governance practices and mapping them onto sociological dimensions of legitimacy. They use case studies from social‑media ranking, search algorithms, and state administrative automation to illustrate where legitimacy breaks down and how legal concepts can be applied without conducting empirical experiments.

## Results  
The analysis shows that technical performance gains do not translate into public trust because audiences lack transparency and are subject to unaccountable authority. Thin legality improves perceived fairness through consistent publicity but remains superficial without thick legality, which ensures rules are publicly authored. The three principles — integration (placing AI rule‑setting in established institutions), familiarity (presenting rules in locally credible formats), and contestation (providing real remedies) — are presented as feasible pathways to restore legitimacy.

## Significance  
This work matters because it shifts focus from technical alignment to the normative legitimacy of AI governance, offering policymakers a roadmap for law that can actually secure public acceptance rather than merely checking boxes. By distinguishing thin and thick legality, it provides concrete criteria for evaluating existing regulations and suggests reforms that embed democratic participation.

## Related Concepts  
- Legitimacy (sociological belief in authority)  
- Alignment (technical objective matching values)  
- Thin legality (formal compliance, publicity)  
- Thick legality (public authorship of rules)  
- Opacity, private power, administrative automation