数据驱动型立法：以新质生产力赋能法治现代化的理论与实践

Data-Driven Legislation: Theory and Practice of Empowering the Modernization of the Rule of Law with New Quality Productive Forces

摘要: 新质生产力的发展推动社会生产关系发生了深刻变革，也对法治现代化提出全新要求，传统经验主导的渐进式立法已经难以满足数字时代发展复杂需求。在此背景下，数据驱动型立法作为新质生产力与法治建设深度融合的产物应运而生。本文立足于立法模式的转型，根据新质生产力理论、民主正当性理论及更精细化的实质正义理论阐释数据驱动型立法的必要性与合理性，结合《生成式人工智能服务管理暂行办法》的立法实践，系统阐释数据驱动在立法前的需求识别、立法中的草案起草与影响评估、立法后的效果监测与动态优化全流程的应用机制。同时，本文分析了数据驱动型立法面临的数据质量失真、立法者的技术依赖、数字鸿沟导致的公平性缺失等潜在风险，并提出了建立数据治理体系、明确立法者的主体地位、完善数据民主补偿机制等应对举措。研究表明，数据驱动型立法并非摒弃传统立法经验，而是通过技术赋能实现立法从“事后应对”向“前瞻布局”的转变，其未来将在技术融合、制度完善与应用范围拓展中持续发展，为中国式现代化提供坚实法治保障。

Abstract: The development of new quality productive forces has triggered profound transformations in social production relations and set brand-new requirements for the modernization of the rule of law. Traditional incremental legislation dominated by experience can hardly meet the complex developmental demands of the digital era. Against this backdrop, data-driven legislation has emerged as an outcome of the in-depth integration between new quality productive forces and the construction of the rule of law. From the perspective of the transformation of legislative models, this paper illustrates the necessity and rationality of data-driven legislation based on the theories of new quality productive forces, democratic legitimacy and refined substantive justice. Drawing on the legislative practice of the Interim Measures for the Administration of Generative Artificial Intelligence Services, it systematically elaborates the application mechanism of data-driven approaches throughout the whole legislative process: demand identification prior to legislation, draft formulation and impact assessment during legislation, as well as effect monitoring and dynamic optimization after legislation. Meanwhile, this paper analyzes potential risks confronting data-driven legislation, including distorted data quality, legislators’ technological dependence, and lack of fairness stemming from the digital divide. Corresponding countermeasures are proposed, such as establishing a data governance system, clarifying legislators’ principal status, and improving the compensatory mechanism for data-enabled democracy. The study demonstrates that data-driven legislation does not discard traditional legislative experience. Instead, it leverages technological empowerment to shift legislation from “ex-post response” to “forward-looking planning”. It will keep evolving through technological integration, institutional improvement and expanded application scope, so as to deliver solid legal safeguards for Chinese-path modernization.