As societal challenges grow more complex, access to data for public interest use is paradoxically becoming more constrained. This emerging data winter is not simply a matter of scarcity, but of shrinking legitimate and trusted pathways for responsible data reuse. Concerns over misuse, regulatory uncertainty, and the competitive race to train AI systems have concentrated data access among a few actors while raising costs and inhibiting collaboration. Prevailing data governance models, focused on compliance, risk management, and internal control, are necessary but insufficient. They often result in data that is technically available yet practically inaccessible, legally shareable yet institutionally unusable, or socially illegitimate to deploy. This paper proposes strategic data stewardship as a complementary institutional function designed to systematically, sustainably, and responsibly activate data for public value. Unlike traditional stewardship, which tends to be inwardlooking, strategic data stewardship focuses on enabling cross sector reuse, reducing missed opportunities, and building durable, ecosystem-level collaboration. It outlines core principles, functions, and competencies, and introduces a practical Data Stewardship Canvas to support adoption across contexts such as data collaboratives, data spaces, and data commons. Strategic data stewardship, the paper argues, is essential in the age of AI: it translates governance principles into practice, builds trust across data ecosystems, and ensures that data are not only governed, but meaningfully mobilized to serve society.

翻译：随着社会挑战日益复杂，公共利益用途的数据获取却矛盾地变得更加受限。这一新兴的"数据寒冬"不仅是数据稀缺的问题，更是负责任数据再利用的合法可信途径不断萎缩的困境。对数据滥用、监管不确定性以及人工智能系统训练竞争的担忧，使得数据访问权集中在少数行为者手中，同时提高了成本并抑制了协作。当前主流的数据治理模式侧重于合规、风险管理和内部控制，这些虽属必要但尚不充分。它们常常导致数据在技术上可用却实际难以获取、法律上可共享却因制度限制无法使用，或在社会层面缺乏部署的正当性。本文提出战略性数据管理作为一种补充性制度功能，旨在系统化、可持续且负责任地激活数据以创造公共价值。与传统的内向型管理模式不同，战略性数据管理侧重于实现跨部门再利用、减少错失机会，并建立持久的生态系统级协作。本文阐述了其核心原则、功能与能力要求，并引入实用的"数据管理画布"工具，以支持在数据协作体、数据空间和数据公地等不同情境中的应用。本文论证指出，在人工智能时代，战略性数据管理至关重要：它将治理原则转化为实践，在数据生态系统中建立信任，并确保数据不仅得到治理，更能被有效动员以服务社会。