Scholarlink

<p>Contemporary AI governance faces a structural dilemma: hard laws are too rigid to adapt to rapid technological change, while soft laws lack enforceability. Within this regulatory impasse, trust has emerged as a critical yet under-theorized concept. This paper reconceptualizes trust as a meso-level legal and governance principle, drawing on sociological theories of institutional trust and systemic complexity. It proposes a three-dimensional framework for institutionalizing trust in AI governance, comprising: (1) normative values—honesty, justice, protection, and loyalty; (2) governance mechanisms—regulatory pyramids, reputation systems, and certification schemes; and (3) institutional embedding through law, policy, and organizational design. Further, the paper introduces the Trust-State Response Matrix, a model that maps varying trust conditions to tailored regulatory strategies. This work contributes by (1) developing a “third way” between hard and soft law through institutionalized trust, and (2) advancing trust as both normative foundation and practical mechanism for sustainable AI governance.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Institutionalizing trust in AI governance: from ethical principles to legal design

  • 期刊论文

  • 发表日期：

    2026年2月22日

查看全文
  • Xinzi Lyu

摘要

Contemporary AI governance faces a structural dilemma: hard laws are too rigid to adapt to rapid technological change, while soft laws lack enforceability. Within this regulatory impasse, trust has emerged as a critical yet under-theorized concept. This paper reconceptualizes trust as a meso-level legal and governance principle, drawing on sociological theories of institutional trust and systemic complexity. It proposes a three-dimensional framework for institutionalizing trust in AI governance, comprising: (1) normative values—honesty, justice, protection, and loyalty; (2) governance mechanisms—regulatory pyramids, reputation systems, and certification schemes; and (3) institutional embedding through law, policy, and organizational design. Further, the paper introduces the Trust-State Response Matrix, a model that maps varying trust conditions to tailored regulatory strategies. This work contributes by (1) developing a “third way” between hard and soft law through institutionalized trust, and (2) advancing trust as both normative foundation and practical mechanism for sustainable AI governance.