通过AI文档提升治理成效：弥合理论与实践的差距
Improving governance outcomes through AI documentation: Bridging theory and practice
AI总结：
该研究通过分析37项AI文档框架与22项实证研究，明确文档提升治理的路径，揭示从业者面临的资源不足、流程整合差等挑战，并提出组织设计文档的关键考量，为AI治理的文档实践与未来研究提供参考。
摘要：
文档在AI系统的外部问责与内部治理中均发挥着关键作用。尽管目前已有诸多针对AI数据、模型、系统及方法的文档编制提案，但这些实践如何强化治理，以及从业者和组织在文档工作中面临的挑战，仍未得到充分探究。本文分析了37项已提出的文档框架，以及22项评估其应用情况的实证研究。我们明确了文档可提升治理的若干路径，即“变革理论”，包括向利益相关方告知AI风险与应用场景、促进协作、推动伦理审议，以及支持最佳实践落地。但实证结果显示，从业者面临诸多重大挑战，例如激励与资源不足、结构性与组织层面的沟通障碍、伦理行动面临的人际及组织约束，以及与现有工作流整合不佳等。这些挑战往往阻碍了文档潜在效益的实现。我们还强调了组织设计文档方案时的关键考量，例如确定适宜的详细程度，以及在流程中平衡自动化应用。最后，我们讨论了未来研究可如何拓展本研究的发现，例如探索支持通用模型治理的文档方法，以及多种透明度与文档方法如何协同提升治理成效。
英文摘要：
Documentation plays a crucial role in both external accountability and internal governance of AI systems. Although there are many proposals for documenting AI data, models, systems, and methods, the ways these practices enhance governance as well as the challenges practitioners and organizations face with documentation remain underexplored. In this paper, we analyze 37 proposed documentation frameworks and 22 empirical studies evaluating their use. We identify several pathways or "theories of change" through which documentation can enhance governance, including informing stakeholders about AI risks and applications, facilitating collaboration, encouraging ethical deliberation, and supporting best practices. However, empirical findings reveal significant challenges for practitioners, such as insufficient incentives and resources, structural and organizational communication barriers, interpersonal and organizational constraints to ethical action, and poor integration with existing workflows. These challenges often hinder the realization of the possible benefits of documentation. We also highlight key considerations for organizations when designing documentation, such as determining the appropriate level of detail and balancing automation in the process. We conclude by discussing how future research can expand on our findings such as by exploring documentation approaches that support governance of general-purpose models and how multiple transparency and documentation methods can collectively improve governance outcomes.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
