ESG视角下人工智能大模型风险识别与治理模型

Artificial intelligence foundation model risk identification and governance model from the ESG perspective

人工智能大模型应用生态快速拓展,环境、社会和治理面临新的挑战和机遇,探索构建大模型发展风险的治理框架,对于推进人工智能健康可持续发展具有重要的理论价值和实践意义。文章基于ESG (环境—社会—治理)和人工智能治理相关理论,分析ESG视角下人工智能大模型发展效益、典型风险,进而构建人工智能大模型风险治理框架和实施策略。研究认为技术治理和制度创新双管齐下是应对大模型发展风险的必要手段。遵循“切入视角→风险识别→治理机制→治理策略”的逻辑,文章探索性地提出了大模型风险治理框架,并根据模型风险影响周期和影响范围,提出“放开、管控、试点、迭代”4类差异化策略。

The application ecology of artificial intelligence foundation model is rapidly expanding.The environment,society,and governance are facing new challenges and opportunities.Exploring the construction of a governance framework for the development risks of foundation model has important theoretical value and practical significance for promoting the healthy and sustainable development of artificial intelligence.Based on the theories of ESG and artificial intelligence governance,this study analyzes the development benefits and typical risks of foundation model from the perspective of ESG and then constructs a risk governance framework and implementation strategies for artificial intelligence foundation models.This study shows that a dual approach of technological governance and institutional innovation is necessary to address the development risks of foundation models.Following the logic of“entry perspective→risk identification→governance mechanism→governance strategy”,the study exploringly proposes the risk governance framework of foundation model.Based on the impact cycle and scope of the foundation model risk,this study proposes four differentiated governance strategies,including“release,control,pilot,and iteration”.