查詢結果分析
來源資料
頁籤選單縮合
| 題 名 | 人工智慧在全球氣候治理的機會與風險=Opportunities and Risks of Artificial Intelligence in Global Climate Governance |
|---|---|
| 作 者 | 楊惟任; | 書刊名 | 中國行政評論 |
| 卷 期 | 32:1 2026.03[民115.03] |
| 頁 次 | 頁1-31 |
| 分類號 | 312.83 |
| 關鍵詞 | 氣候變遷; 人工智慧; 全球氣候治理; 巴黎協定; 社會技術想像; Climate change; Artificial intelligence; AI; Global climate governance; Paris Agreement; Sociotechnical imaginaries; |
| 語 文 | 中文(Chinese) |
| DOI | 10.6635/cpar.202603_32(1).0001 |
| 中文摘要 | 在氣候危機持續加劇與數位科技迅速發展的背景下,人工智慧漸漸成為全球氣候治理的重要支撐,多數國際組織將其視為強化氣候行動,推進並落實《巴黎協定》的關鍵解方。不過,既有研究多著重於人工智慧在氣候減緩與調適的技術效能,對其背後所蘊含的權力關係、規範假設與政治後果,仍缺乏系統性的討論。為弭補此一缺口,本文跳脫工具理性取向,採取「社會技術想像」理論視角,探究國際組織如何於全球氣候治理場域建構人工智慧的角色與正當性,並分析此種想像如何形塑氣候治理的問題界定、知識生產與資源分配。本文採質性研究設計,結合文獻分析與批判論述分析,檢視《聯合國氣候變遷綱要公約》、政府間氣候變遷專門委員會、聯合國環境規劃署、世界氣象組織、世界銀行、經濟合作暨發展組織、世界經濟論壇等國際機構之政策文件。研究發現,這些機構普遍建構高度一致的技術樂觀敘事,將人工智慧描繪為中立、高效率且具普遍適用性的治理工具,並傾向把氣候危機理解為可透過資料化與演算法優化加以處理的技術性問題。然而,此一社會技術想像掩蓋人工智慧在能耗、資料不平等、權力集中等結構性風險,並加劇全球南北在技術能力、知識生產與氣候資源配置的不對稱。本文主張,人工智慧並非價值中立的工具,其在全球氣候治理的擴散應被理解為嵌入既有權力結構的政治過程,若欲避免治理陷入技術依賴,並抑制分配不平等的擴大,必須將人工智慧置於更具民主性、公平性與環境問責的治理框架之中。 |
| 英文摘要 | Against the backdrop of an intensifying climate crisis and the rapid advancement of digital technologies, artificial intelligence (AI) is becoming an increasingly pivotal pillar of global climate governance. Many international organizations view AI as a key solution for strengthening climate action and advancing the implementation of the Paris Agreement. Existing scholarship, however, has largely emphasized AI's technical efficacy for climate mitigation and adaptation, while offering limited systematic discussion of the power relations, normative assumptions, and political consequences embedded in its deployment. To address this gap, this article moves beyond an instrumental-rational perspective and adopts the theoretical lens of sociotechnical imaginaries to examine how international organizations construct AI's role and legitimacy within the arena of global climate governance, and how such imaginaries shape problem framing, knowledge production, and resource allocation. Methodologically, the study employs a qualitative research design, combining literature review with critical discourse analysis to analyze policy documents produced by major international institutions, including the United Nations Framework Convention on Climate Change, the Intergovernmental Panel on Climate Change, the United Nations Environment Program, the World Meteorological Organization, the World Bank, the Organization for Economic Co-operation and Development, and the World Economic Forum. The findings reveal that these institutions commonly advance a highly convergent techno-optimistic narrative, portraying AI as a neutral, highly efficient, and universally applicable governance instrument. They also tend to frame the climate crisis as a technical problem amenable to datafication-the transformation of social and environmental phenomena into quantifiable data and algorithmic optimization, which refers to the process of improving decision-making and resource allocation through automated computational procedures. Yet the sociotechnical imaginary obscures structural risks associated with AI, such as energy consumption, data inequalities, and the concentration of power, and exacerbates asymmetries between the Global North and the Global South in technological capacity, knowledge production, and the allocation of climate-related resources. AI is not a value-neutral tool; its role in global climate governance is shaped by existing power structures. To prevent overreliance on technology and increased inequities, AI should be embedded in governance frameworks that prioritize democracy, equity, and environmental accountability. |
本系統中英文摘要資訊取自各篇刊載內容。