| 注册
首页|期刊导航|当代经济管理|场景驱动视角下的大模型治理策略与模式选择

场景驱动视角下的大模型治理策略与模式选择

李少帅

当代经济管理2025,Vol.47Issue(5):27-33,7.
当代经济管理2025,Vol.47Issue(5):27-33,7.DOI:10.13253/j.cnki.ddjjgl.2025.05.004

场景驱动视角下的大模型治理策略与模式选择

Governance Strategies and Model Selection of Large Models from a Scenario-driven Perspective

李少帅1

作者信息

  • 1. 清华大学,北京 100084
  • 折叠

摘要

Abstract

Technological innovation and application development in the field of large model,represented by ChatGPT,DeepSeek,are exceptionally active,triggering a new wave of development in general artificial intelligence.The large model has shown broad development prospects in e-commerce,financial advisory and other scenarios,but it has also triggered new governance challenges.Exploring the governance strategies and models of large model has important theoretical value and practical significance for promoting high-quality and sustainable development.This paper starts from scenario-driven perspective and divides the application context of large model into three types:commercial applications,mixed public and commercial applica-tions,and public services.It explores the positive and negative external impact of large models in specific scenarios,and proposes specific governance strategies for creating technological value,preventing and resolving risks,and building technological trust in specific context from the perspective of de-velopment and regulation.The study further in-depth analyzes the governance models of large models based on scenario adaptation,including six gov-ernance models such as adaptive governance and experimental governance,thereby transforming the above governance strategies into concrete actions in specific scenarios.The research provides theoretical basis and practical guidance for more precise and effective governance of large models in specific scenarios.

关键词

场景驱动/大模型/场景属性/治理策略/治理模式

Key words

scenario-driven/large model/scenario properties/governance strategy/governance model

分类

管理科学

引用本文复制引用

李少帅..场景驱动视角下的大模型治理策略与模式选择[J].当代经济管理,2025,47(5):27-33,7.

基金项目

科技创新2030—"新一代人工智能"重大项目"新一代人工智能风险防范与治理手段研究"(2023ZD0121700) (2023ZD0121700)

国家资助博士后研究人员计划(GZC20240796). (GZC20240796)

当代经济管理

OA北大核心

1673-0461

访问量0
|
下载量0
段落导航相关论文