Rethinking Higher Education/Chapter 11/en-zh

Green Digital Education: Sustainability, Digital Sobriety, and Environmental Awareness in EU and Chinese Universities

绿色数字教育：欧盟与中国高校的可持续发展、数字节制与环保意识

Martin Woesler 吴漠汀

'Abstract' 摘要

The digital transformation of higher education carries a hidden environmental cost that is rarely acknowledged in educational policy discourse. Data centres consumed 415 terawatt-hours of electricity in 2024 — 1.5 percent of global electricity demand — and are projected to reach 945 TWh by 2030, growing four times faster than any other sector. Training a single large language model such as GPT-3 produces approximately 552 metric tons of CO2 equivalent, comparable to the annual emissions of 121 average American households. Annual digital content consumption generates 229 kg of CO2 equivalent per user, approximately 3–4 percent of per capita anthropogenic greenhouse gas emissions. Yet the educational technology sector has largely escaped scrutiny for its environmental footprint, even as universities expand their digital infrastructure under the banners of innovation and accessibility. This article examines the tension between digital education and environmental sustainability through a systematic comparison of European Union and Chinese approaches. The EU has developed the GreenComp framework for sustainability competences and is beginning to address „digital sobriety„ — the principle of minimizing unnecessary digital consumption — as an educational goal. China has pursued „ecological civilization education“ as a framework that integrates environmental awareness with broader civilizational goals, while simultaneously undertaking the world’s largest expansion of digital educational infrastructure. We argue that both systems face an „AI energy paradox„ — the deployment of artificial intelligence in education simultaneously promises to enhance sustainability awareness and contributes substantially to environmental degradation — and that neither has yet developed an adequate response.

高等教育的数字化转型暗藏着一项隐性环境成本，而这一点在教育政策讨论中却极少被提及。2024年，全球数据中心耗电量达415太瓦时，占全球电力总需求的1.5%；预计到2030年耗电量将增至945太瓦时，增速是其他任何行业的四倍。 训练单个大型语言模型（如GPT-3）约产生552公吨二氧化碳当量，相当于121户普通美国家庭的年碳排放总量。用户每年的数字内容消费人均产生229公斤二氧化碳当量，约占人均人为温室气体排放量的3%—4%。然而，即便各大高校以创新与教育普惠为名不断扩建数字基础设施，教育科技行业自身的环境足迹却始终未受到应有审视。 本文通过系统对比欧盟与中国的发展路径，剖析数字教育与环境可持续发展之间的内在矛盾。欧盟已构建GreenComp可持续发展能力框架，并开始将数字节制——最大限度减少非必要数字消耗的原则——确立为教育目标。中国以生态文明教育为发展框架，将环保意识融入更宏观的文明建设目标，同时推进全球规模最大的数字教育基础设施建设。 本文认为，两大高等教育体系均面临人工智能能源悖论：人工智能在教育领域的应用，一方面有望提升公众的可持续发展意识，另一方面又在极大程度上加剧了环境退化，而目前双方均未拿出完善有效的应对方案。 Keywords: green digital education, sustainability, digital sobriety, ecological civilization, carbon footprint, data centres, AI energy paradox, GreenComp, higher education, EU-China comparison 关键词： 绿色数字教育、可持续发展、数字节制、生态文明、碳足迹、数据中心、人工智能能源悖论、欧洲可持续能力框架、高等教育、中欧比较

'1. Introduction' 1. 引言

The environmental sustainability of digital education is a topic that most educational technologists would prefer not to discuss. The digital transformation of higher education has been driven by powerful narratives of progress: AI-personalized learning, immersive virtual reality, global connectivity, and institutional efficiency. These narratives are not wrong — the companion chapters in this anthology document genuine educational benefits from digital technologies (Woesler, this volume). But they are incomplete, because they systematically ignore the material basis of digital education: the servers, networks, devices, and energy systems that make it possible, and the environmental consequences of operating them at scale. 数字教育的环境可持续性，是绝大多数教育技术从业者刻意回避的话题。高等教育数字化转型，始终被人工智能个性化学习、沉浸式虚拟现实、全球互联共享、院校治理效能提升等进步叙事所主导。这些叙事本身并无谬误——本文集同系列章节也证实了数字技术确实能为教育带来实质增益（韦斯勒，本卷）。 但这类叙事存在明显短板：它们刻意回避数字教育赖以运转的物质基础——服务器、网络设备、终端硬件与能源供给体系，以及规模化运行所带来的各类环境代价。

The numbers are sobering. The International Energy Agency reports that data centres consumed 415 terawatt-hours of electricity in 2024, representing 1.5 percent of global electricity demand. This figure is projected to reach 945 TWh by 2030 — more than Japan’s total electricity consumption — with growth rates of approximately 15 percent per year, four times faster than all other sectors combined (IEA 2025). The carbon emissions of the major technology companies have surged in parallel: Google’s greenhouse gas emissions rose 48 percent between 2019 and 2024, while Microsoft’s grew 29 percent since 2020, with data centre energy consumption identified as the primary driver (NPR 2024). An analysis of corporate sustainability reports suggests that actual data centre emissions may be 7.62 times higher than reported, due to accounting practices that count renewable energy certificates as equivalent to actual renewable energy consumption (Le Goff 2025).

数据触目惊心。国际能源署报告显示，2024年全球数据中心耗电415太瓦时，占全球电力需求1.5%；预计2030年将攀升至945太瓦时，超过日本全国年用电量，年均增速约15%，是其余所有行业增速总和的四倍（IEA 2025）。头部科技企业碳排放量同步飙升：2019至2024年谷歌温室气体排放量上涨48%，微软自2020年起增幅达29%，核心诱因均为数据中心能耗激增（NPR 2024）。另有企业可持续报告分析指出，受可再生能源证书记账规则影响（将绿证等同于实际绿电消耗），数据中心实际碳排放或比公示数据高出7.62倍（Le Goff 2025）。

For higher education, these figures have direct relevance. Universities are among the largest institutional consumers of digital infrastructure, operating learning management systems, research computing clusters, videoconferencing platforms, and — increasingly — AI-powered educational tools. Yet the environmental footprint of this digital infrastructure is almost never included in university sustainability assessments. Williamson, Hogan, and Selwyn (2025), in a chapter for a Springer volume on critical EdTech studies, argue that the environmental impact of educational technology platforms is „persistently overlooked“ in university carbon calculations, drawing attention to the globally distributed, energy-intensive IT infrastructure in which universities are enmeshed through their EdTech partnerships. 对高等院校而言，这些数据具有直接现实关联。高校是数字基础设施最大的机构用户群体之一，日常运营教学管理系统、科研计算集群、视频会议平台，如今更逐步普及人工智能教学工具。但这类数字基础设施的环境足迹，几乎从未纳入高校可持续发展评估体系。 威廉姆森、霍根与塞尔温（2025）在施普林格出版社教育技术批判研究文集中指出，高校碳核算长期忽视教育科技平台的环境影响；高校通过校企合作深度绑定全球高耗能信息技术基建，这一现实始终未被重视。

This article examines how the European Union and China — the two largest higher education systems by enrollment — are addressing (or failing to address) the environmental dimension of their digital education strategies. We compare the EU’s emerging framework of sustainability competences and digital sobriety with China’s ecological civilization education, assessing each for its capacity to confront the environmental costs of the digital infrastructure on which modern education increasingly depends. 本文以全球在校生规模最大的两大高等教育体系——欧盟与中国为研究对象，探析双方在数字教育战略中对环境维度的应对现状（及缺失之处）。对比欧盟新兴的可持续发展能力框架与数字节制理念、中国生态文明教育体系，评估两大框架能否有效化解现代教育高度依赖数字基建所产生的环境成本问题。

'2. The Carbon Footprint of Educational Technology' 2. 教育科技的碳足迹

'2.1 Data Centres and AI Training'

2.1 数据中心与人工智能模型训练

The energy consumption of digital infrastructure has two main components: the operational energy of data centres (including cooling, which can account for 30–40 percent of total energy use) and the embodied energy of hardware production and disposal. For AI systems specifically, a third component — the energy consumed during model training — has become increasingly significant. 数字基础设施能耗主要分为两部分：一是数据中心运行能耗（含制冷能耗，占总能耗30%—40%）；二是硬件生产与报废的隐含能耗。对人工智能系统而言，模型训练能耗已成为第三大核心能耗来源，占比日益凸显。

Patterson and colleagues (2021), in a study by researchers at Google and UC Berkeley, estimated that training GPT-3 produced approximately 552 metric tons of CO2 equivalent and consumed 1,287 megawatt-hours of electricity. Strubell, Ganesh, and McCallum (2019), in the paper that first drew wide attention to the carbon cost of training large language models, demonstrated that training a single large NLP model can emit as much carbon as five automobiles over their entire lifetimes. These figures have grown substantially with the development of larger models: GPT-4 and its successors consume orders of magnitude more energy, though precise figures are not publicly disclosed. 帕特森等人（2021）联合谷歌与加州大学伯克利分校开展研究，测算得出训练GPT-3模型约产生552公吨二氧化碳当量，耗电1287兆瓦时。斯特鲁贝尔、加内什与麦卡勒姆（2019）率先揭示大语言模型训练的碳排放代价，研究表明：训练单个大型自然语言处理模型的碳排放量，相当于五辆汽车全生命周期碳排放总和。随着大模型迭代升级，这一数值大幅攀升：GPT-4及后续模型能耗高出数个数量级，具体数据并未公开披露。

The water footprint of AI is equally concerning. Li and colleagues (2023), in a study published in Communications of the ACM, estimate that training GPT-3 in Microsoft’s US data centres directly evaporated 700,000 litres of freshwater. Global AI water demand is projected to reach 4.2–6.6 billion cubic metres by 2027 — more than the total annual water withdrawal of four to six Denmarks. De Vries (2025), writing in the Cell Press journal Patterns, estimates the AI industry’s 2025 carbon footprint at 32.6–79.7 million metric tons of CO2, comparable to the total emissions of New York City, with a water footprint of 312.5–764.6 billion litres. 人工智能的水足迹同样不容忽视。李等人（2023）在《美国计算机学会通讯》刊发研究，测算微软美国数据中心训练GPT-3直接蒸发淡水70万升。全球人工智能行业用水量预计2027年达42亿—66亿立方米，相当于4至6个丹麦全年总取水量。弗里斯（2025）在细胞出版社《Patterns》期刊发文估算，2025年人工智能行业碳足迹达3260万—7970万公吨二氧化碳当量，与纽约市年碳排放总量持平；水足迹高达3125亿—7646亿升。

For universities, these aggregate figures translate into institutional responsibility. Every time a student uses a cloud-based AI writing assistant, submits work to an AI-powered plagiarism detector, or engages with an AI tutoring system, the university’s digital infrastructure generates emissions that are invisible to the user but cumulatively significant. Xiao and colleagues (2025), writing in Nature Sustainability, argue that the US AI industry is unlikely to meet net-zero targets by 2030 without substantial reliance on „highly uncertain carbon offset and water restoration mechanisms.“

对高校而言，宏观数据最终落脚为院校责任。学生每使用一次云端AI写作助手、提交作业至AI查重系统、参与人工智能辅导课程，高校数字基建都会产生用户无法感知、但累积效应显著的碳排放。肖等人（2025）在《自然·可持续发展》发文指出，若过度依赖不确定性极高的碳抵消与水资源修复机制，美国人工智能行业很难在2030年前实现净零排放目标。

'2.2 Digital Content Consumption'

2.2 数字内容日常消费

Beyond AI training, the routine digital activities of education carry their own environmental costs. Istrate and colleagues (2024), in a study published in Nature Communications, estimate that annual global average digital content consumption — web browsing, social media, video and music streaming, and videoconferencing — generates 229 kg of CO2 equivalent per user per year, approximately 3–4 percent of per capita anthropogenic greenhouse gas emissions. Under a 1.5 degrees Celsius warming scenario, this could account for approximately 40 percent of the per capita carbon budget. 除人工智能模型训练外，教育场景常态化数字活动同样产生可观环境成本。伊斯特拉特等人（2024）于《自然·通讯》刊发研究：全球用户年均网页浏览、社交媒体、音视频流媒体、视频会议等数字内容消费，人均产生229公斤二氧化碳当量，占人均人为温室气体排放的3%—4%。在全球升温控制1.5摄氏度的情景下，该占比或将达到人均碳预算的40%。

For universities, the implications are significant. A single semester of online course delivery for thousands of students involves substantial video streaming, file sharing, and platform interaction. Caird and Lane (2024), writing in the Future Healthcare Journal, note that while digital learning generally has a lower carbon footprint than face-to-face instruction when travel is factored in — travel to in-person conferences can produce 1,000 times more CO2 than virtual alternatives — the comparison is less favorable when the full lifecycle costs of digital infrastructure are included. 对高校而言，其影响尤为显著。数千名学生整学期线上授课，伴随海量视频流媒体播放、文件传输共享、平台交互使用。凯尔德与莱恩（2024）在《未来医疗期刊》指出：若计入通勤成本，纯数字化学习碳足迹通常低于线下面授（线下学术会议通勤碳排放可达线上模式的1000倍）；但如果纳入数字基建全生命周期成本，这一优势将大幅弱化。

'2.3 E-Waste and the Hardware Lifecycle'

2.3 电子垃圾与硬件全生命周期

The environmental cost of digital education extends beyond energy consumption to include the physical devices on which it depends. The accelerating pace of hardware replacement in educational institutions — driven by software requirements, institutional procurement cycles, and the planned obsolescence of consumer electronics — generates a growing stream of electronic waste that is rarely included in discussions of sustainable education. 数字教育的环境代价不止能耗消耗，更涵盖其赖以运行的实体终端设备。受软件版本迭代、院校采购周期、消费电子计划性淘汰等因素影响，高校硬件更新换代节奏持续加快，电子垃圾排放量逐年攀升，却极少纳入可持续教育议题讨论范畴。

Valai Ganesh and colleagues (2025), in a study published in Scientific Reports, examined 452 electrical products across academic institutions in India and found that 32.1 percent were older than five years and 34.1 percent needed repair or replacement. Their proposed sustainable e-waste management framework demonstrated that on-site recycling can achieve a 90 percent material recovery rate — but such frameworks require institutional investment and commitment that most universities have not yet made. Thao, Hanh, and Huy (2025), in a study of Ho Chi Minh City University of Technology published in the International Journal of Environmental Science and Technology, project that e-waste at a single university campus will increase 1.5 times from 16,792 kg in 2024 to 25,230 kg in 2034, reflecting the growing hardware intensity of digital education. 瓦莱·加内什等人（2025）在《科学报告》发表研究，调研印度高校452台电气设备后发现：32.1%设备使用年限超五年，34.1%设备需维修或更换。 其提出的可持续电子垃圾治理框架显示，现场回收可实现90%物料回收率，但该模式需要高校持续投入资金与制度保障，目前多数院校尚未落地。 陈、韩、辉（2025）针对胡志明市理工大学的研究刊发于《环境科学与技术国际期刊》，测算该校校园电子垃圾将从2024年的16792公斤增至2034年的25230公斤，十年增幅1.5倍，直观反映数字教育硬件密集度持续走高的趋势。

For Chinese and European universities alike, the e-waste problem is compounded by the trend toward institutional tablet and laptop programs, one-to-one device initiatives, and the regular replacement of smart classroom equipment. When a university deploys thousands of tablets for a digital learning initiative, the educational benefit may be genuine — but the environmental cost of manufacturing, operating, and eventually disposing of those devices is rarely calculated. The concept of digital sobriety, discussed in the following section, offers a framework for addressing this gap. 中欧高校均面临电子垃圾加剧的困境：院校批量采购平板、笔记本一人一机项目、智慧教室设备常态化换新，进一步放大污染压力。 高校推行数字化学习项目批量部署数千台终端，虽能收获实实在在的教学效益，但设备生产、运行、最终报废的全周期环境成本，始终缺乏量化核算。 下一节提出的数字节制理念，恰好为填补这一治理空白提供理论框架。

'3. Digital Sobriety as Educational Goal'

3.作为教育目标的数字节制

'3.1 Origins and Definition' 3.1 起源与定义

The concept of „digital sobriety„ (sobriété numérique) originated with the French think tank The Shift Project, whose 2019 report „Lean ICT: Towards Digital Sobriety“ defined it as the principle of buying the least powerful equipment possible, changing devices as rarely as possible, and reducing unnecessary energy-intensive digital uses. The report estimated that ICT energy consumption was increasing at 9 percent annually and argued that a sobriety approach could limit growth to 1.5 percent (The Shift Project 2019). 数字节制（法语：sobriété numérique）由法国智库“转型项目研究所”首创。该机构2019年发布报告《精简信息通信技术：迈向数字节制》，将其定义为：选用性能适配、不过度冗余的设备，尽量延长终端更换周期，减少非必要、高耗能的数字使用行为。 报告测算全球信息通信技术能耗年均增速达9%，而推行数字节制理念可将增速控制在1.5%以内（转型项目研究所，2019）。

In education, digital sobriety has gained recognition through UNESCO’s 2024 decision to award the King Hamad Bin Isa Al-Khalifa Prize for ICT in Education to the Belgian EducoNetImpact initiative, which promotes sustainable digital practices in schools. Approximately 1,000 teachers now use its materials (UNESCO 2024). This recognition signals that the international educational community is beginning to acknowledge the environmental dimension of educational technology — though the scale of the response remains modest relative to the scale of the problem. 在教育领域，数字节制已获得联合国教科文组织认可。2024年教科文组织将哈马德·本·伊萨·阿勒哈利法信息技术教育奖，授予比利时EducoNetImpact项目——该项目致力于在中小学推广可持续数字使用范式，目前已有约1000名教师使用其教学资源（联合国教科文组织，2024）。 这一认可标志着国际教育界开始正视教育科技的环境维度，但相较于问题规模，全球应对力度仍显不足。

'3.2 The EU Framework: DigComp and GreenComp' 3.2 欧盟框架：DigComp 与 GreenComp

The EU’s approach to sustainability in digital education draws on two complementary frameworks. The DigComp 2.2 framework (Vuorikari, Kluzer, and Punie 2022) incorporates sustainability-related examples within its five competence areas, addressing the environmental implications of digital technology use. However, digital sobriety is not explicitly named as a competence dimension within DigComp 2.2 — a gap that suggests the framework’s development has not fully caught up with the emerging environmental concerns. 欧盟数字教育可持续发展依托两大互补框架：《公民数字能力框架DigComp 2.2》（沃里卡里、克鲁泽、普尼，2022）在五大能力维度中融入可持续发展相关案例，关注数字技术使用的环境影响。但数字节制并未被明确列为DigComp 2.2的能力维度，反映出该框架更新滞后于当下新兴的环境治理诉求。

The GreenComp framework (Bianchi, Pisiotis, and Cabrera Giraldez 2022), published by the EU Joint Research Centre, provides a complementary framework with four competence areas: embodying sustainability values, embracing complexity in sustainability, envisioning sustainable futures, and acting for sustainability. Its 12 competences are designed to be non-prescriptive reference points for learning schemes across formal and informal education. 欧盟联合研究中心发布的《GreenComp可持续发展能力框架》（比安基、皮西奥蒂斯、卡布雷拉·希拉尔德斯，2022）形成互补，包含四大能力板块：践行可持续价值观、理解可持续发展复杂性、构想可持续未来、践行可持续行动。框架下设12项核心能力，作为正规与非正规教育课程设计的非指令性参考标准。

The GreenSCENT project (Horizon 2020, 2021–2024) has tested the practical application of Green Deal topics in approximately 45 schools and universities across the EU, creating the ECCEL — a European „driving licence“ for climate and environmental competences (European Commission 2021–2024). McDonagh, Caforio, and Pollini’s (2024) edited volume, „The European Green Deal in Education,“ published by Routledge, provides case studies from the project, documenting the first published applications of Green Deal topics in classroom settings. 欧盟地平线2020计划资助的GreenSCENT项目（2021—2024），在欧盟约45所中小学及高校落地测试绿色新政相关教学实践，推出欧洲气候与环境能力认证（ECCEL），相当于欧洲版“环境素养驾照”（欧盟委员会，2021—2024）。麦克多纳、卡福利奥、波利尼（2024）由劳特利奇出版社主编出版《教育中的欧洲绿色新政》论文集，收录该项目典型案例，记录绿色新政理念首次规模化落地课堂教学的实践成果。

Calis and colleagues (2025), in a study of 896 pre-service teachers published in Humanities and Social Sciences Communications, found only a „moderate level“ of digital carbon footprint awareness. Participants showed stronger understanding of electronic device impacts than of data transmission impacts — that is, they understood that manufacturing a laptop has environmental costs but were less aware that streaming a video lecture or using a cloud-based AI tool also generates emissions. Female participants had significantly higher awareness levels than males. This finding is particularly concerning: if teachers themselves are unaware of the environmental costs of digital technology, they cannot be expected to cultivate that awareness in their students. 卡利斯等人（2025）在《人文社会科学通讯》调研896名师范生发现：受访者数字碳足迹认知仅处于中等水平。 师生对电子设备生产环节的环境影响认知较强，但对数据传输、视频网课流媒体、云端AI工具使用的碳排放认知严重不足；女性师范生认知水平显著高于男性。 这一结论值得警惕：若教师群体自身缺乏数字科技环境成本认知，便无法引导学生建立相应环保素养。

At the institutional level, the European University Association’s 2023 report „A Green Deal Roadmap for Universities,“ based on a survey of nearly 400 institutions from 56 higher education systems, found that a large majority of European universities have either incorporated sustainability into their main institutional strategy or developed specific sustainability strategies. However, the majority of institutions called for enhanced funding and more peer-learning opportunities (EUA 2023). The report’s recommendations span public engagement, research, teaching, and campus operations — but the environmental cost of digital infrastructure receives no dedicated treatment, suggesting that even sustainability-committed institutions have not yet integrated digital sobriety into their environmental strategies. 院校层面，欧洲大学协会2023年发布《高校绿色新政路线图》，基于56个高等教育体系近400所院校调研显示：绝大多数欧洲高校已将可持续发展纳入核心办学战略或专项规划，但多数院校呼吁加大经费支持、增加校际经验互鉴（欧洲大学协会，2023）。 报告覆盖公众参与、科研教学、校园运营等维度，却未专门提及数字基建的环境成本，可见即便高度重视可持续发展的欧洲高校，也尚未将数字节制纳入整体环境治理体系。

The growing importance of sustainability reporting in higher education is reflected in the expansion of the Times Higher Education Impact Rankings, which measured universities’ contributions to the UN Sustainable Development Goals. Urbano and colleagues (2025), in an analysis published in the Journal of Cleaner Production, found that the 2024 rankings saw 1,963 participating institutions — a 23 percent increase over the previous year — demonstrating growing institutional commitment to sustainability reporting. However, the rankings’ methodology does not include specific metrics for digital infrastructure emissions, creating a significant blind spot in an otherwise comprehensive assessment framework. 可持续发展报告在高等教育领域的重要性日益凸显，这一点从泰晤士高等教育世界大学影响力排名的扩容中可见一斑。该排名旨在衡量各高校对联合国可持续发展目标的贡献度。 乌尔巴诺及其研究团队于 2025 年在《清洁生产杂志》发表的一项分析研究指出：2024 年该排名共有1963 所高校参与，较上一年度增长 23%，这充分体现了高校机构对可持续发展报告的重视程度正不断提升。但该排名的评估体系并未纳入数字基础设施碳排放的专项衡量指标，使其本已较为完善的评估框架存在明显短板与盲区。

'4. China‘s Approach: Ecological Civilization Education' 4.中国路径：生态文明教育

'4.1 Ecological Civilization as Educational Framework'

4.1 作为教育框架的生态文明

China‘s approach to environmental education is framed not through „sustainability„ in the Western sense but through the concept of „ecological civilization„ (生态文明, shengtai wenming) — a comprehensive framework that integrates environmental protection with economic development, social governance, and civilizational identity. Wang and colleagues (2025), in a chapter for the Springer Handbook of Ecological Civilization, trace the evolution of ecological civilization education as a key project for China’s sustainable development, noting its integration into educational policy at all levels. 中国的环境教育模式并非以西方语境下的 “可持续发展” 为架构，而是立足于生态文明这一核心理念。生态文明是一套综合性体系，将生态环境保护与经济发展、社会治理及文明身份认同融为一体。王等人（2025）在《斯普林格生态文明手册》的专题章节中梳理了生态文明教育的发展脉络，指出其已成为中国可持续发展的重点工程，并全面融入各级教育政策体系之中。

Ecological civilization education differs from European sustainability education in several important respects. First, it is explicitly political: the concept was enshrined in the Chinese Communist Party’s constitution in 2012 and incorporated into the national constitution in 2018, giving it a juridical status that European sustainability frameworks lack. Second, it is comprehensive in scope: ecological civilization encompasses not merely environmental protection but a transformation of the relationship between human civilization and the natural world. Third, it is state-led rather than citizen-oriented: whereas GreenComp empowers individual citizens to make sustainable choices, ecological civilization education positions individuals within a collective project of national transformation. 生态文明教育与欧洲可持续发展教育在多个重要方面存在差异。首先，具有鲜明政治属性：生态文明理念于 2012 年写入党章，2018 年纳入国家宪法，使其拥有欧洲可持续发展框架所不具备的法定地位。其次，覆盖范畴更为全面：生态文明不只局限于生态环境保护，更旨在重塑人类文明与自然之间的关系。第三，以国家主导为主，而非公民本位：欧洲绿色核心素养框架（GreenComp）侧重赋能个体公民做出可持续生活选择，而生态文明教育则将个人置于国家转型发展的集体事业大局之中。

Zhou (2024), in a study published in Social Inclusion, examines the transition from Education for Sustainable Development (ESD) to ecological civilization in China through a climate justice framework. The analysis finds that ecological civilization is heavily political, limited primarily to environmental sustainability (neglecting social and economic dimensions), and that education stakeholders are underrepresented in decision-making processes. These findings suggest that while the framework is ambitious in scope, its top-down implementation may limit its capacity to foster the kind of critical, participatory environmental engagement that the GreenComp framework envisions. 周（2024）在《社会包容》期刊发表的一项研究中，借助气候正义分析框架，考察了中国从可持续发展教育（ESD） 向生态文明教育的转型过程。该研究发现：生态文明具有浓厚的政治属性，其内涵目前主要局限于环境可持续层面，忽视了社会与经济维度；同时，教育领域相关利益主体在决策过程中缺乏充分话语权、代表性不足。上述研究结果表明，尽管生态文明教育框架立意宏大、涵盖范围广泛，但其自上而下的推行模式，可能难以培育出欧盟绿色素养框架（GreenComp）所倡导的批判性、参与式环保实践素养。

Tian and colleagues (2024), in a bibliometric review published in Humanities and Social Sciences Communications analyzing 25 years of Chinese ESD research through LDA topic modelling and social network analysis, identify a „trending-declining“ publication pattern with ample space for expansion. The shift from internationally aligned sustainability goals to a localized, politicized framework under Xi Jinping’s ecological civilization concept is identified as a defining characteristic of Chinese ESD — a trajectory that has both strengths (political commitment, institutional backing) and limitations (reduced critical engagement, limited international comparability). 田等人（2024）在发表于《人文与社会科学传播》期刊的文献计量综述中，通过光谱分析主题建模和社交网络分析分析了中国25年的电子静电现象研究，指出了一种“趋势-下降”的发表模式，并有充足的扩展空间。从国际一致的可持续发展目标转向在习近平生态文明理念下实现本地化、政治化框架，被认为是中国ESD的标志性特征——这一轨迹既有优势（政治承诺、制度支持）也有局限（批判性参与减少，国际可比性有限）。”

'4.2 China‘s Dual Carbon Goals and Higher Education' 4.2 中国的双碳目标与高等教育

A distinctive feature of China‘s approach is the direct integration of national climate targets into higher education policy. In 2022, the Ministry of Education issued a „Work Program for Building a Strong Carbon Peak Carbon Neutral Higher Education Talent Training System“ (加强碳达峰碳中和高等教育人才培养体系建设工作方案), mandating universities to establish new faculties, courses, and vocational programs aligned with China’s dual carbon goals of reaching peak emissions by 2030 and carbon neutrality by 2060. As of 2022, 21 undergraduate programs were directly related to dual-carbon pledges, covering new energy, smart grids, carbon storage, hydrogen energy, and big data for environmental resources (Ministry of Education 2022). 中国发展路径的一大鲜明特色，是将国家气候目标直接融入高等教育政策。2022年，教育部印发《加强碳达峰碳中和高等教育人才培养体系建设工作方案》，要求高校增设相关院系、课程及职业教育项目，对标我国2030年前实现碳达峰、2060年前实现碳中和的双碳目标。截至2022年，全国已有21个本科专业直接对接双碳发展需求，覆盖新能源、智能电网、碳封存、氢能、环境资源大数据等领域（教育部，2022年）。

This policy represents a more direct intervention in curriculum design than anything attempted in the EU, where sustainability education remains largely voluntary and institution-led. The dual carbon education mandate reflects China‘s broader governance philosophy: when the state identifies a strategic priority, universities are expected to align their programs accordingly. Whether this top-down approach produces deeper environmental engagement than the EU’s bottom-up framework of competence development remains an open empirical question. 这项政策对课程设计的干预力度，比欧盟以往任何相关举措都更为直接。在欧盟，可持续发展教育在很大程度上仍属于自愿性质，由各院校自主主导推进。双碳教育强制要求体现了中国更宏观的治理理念：国家一旦确定战略重点，高校便需相应调整自身学科与课程设置。这种自上而下的推行模式，是否能比欧盟自下而上、以能力培养为核心的框架，催生更深层次的环保参与意识，仍是一个有待实证研究解答的问题。

Wang and colleagues (2023), in a study published in the Journal of Cleaner Production, developed a novel LEAP-LCA hybrid methodology for assessing the carbon footprint of a medium-sized Chinese university campus. They found that electricity consumption caused 77 percent of total campus carbon emissions and that proposed carbon reduction measures — photovoltaics, energy efficiency improvements, electrification — could reduce emissions by 97 percent by 2060, with electricity decarbonization alone contributing 64.7 percent of the reduction. These findings suggest that Chinese universities have significant potential for carbon reduction, but that realizing this potential requires infrastructure investment and institutional commitment that cannot be achieved through curriculum reform alone. 王等人（2023年）在《清洁生产杂志》发表的一项研究中，构建了一种全新的LEAP-LCA混合研究方法，用于评估中国一所中型大学校园的碳足迹。研究发现，电力消耗占校园总碳排放的77%；而所提出的降碳措施——光伏发电、能效提升、能源电气化——到2060年可实现97%的减排幅度，其中仅电力脱碳一项就能贡献64.7%的减排量。该研究结果表明，中国高校具备巨大的碳减排潜力，但挖掘这一潜力需要基础设施投入与院校层面的坚定举措，仅依靠课程改革无法实现减排目标。

'4.3 Green Campus Initiatives'

4.3 绿色校园举措

China‘s approach to green digital education embodies a distinctive tension. On one hand, the government is pursuing the world’s largest expansion of digital educational infrastructure — the National Smart Education Platform serving 293 million students, near-universal school broadband, mandatory AI education from September 2025. On the other hand, it is simultaneously promoting ecological civilization education and green campus initiatives. 中国在绿色数字教育方面的举措体现了独特的矛盾关系：一方面，政府正推进全球规模最大的数字教育基础设施建设——国家智慧教育平台覆盖2.93亿学生，实现近乎全民的校园宽带覆盖，并自2025年9月起强制推行人工智能教育；另一方面，同时大力推广生态文明教育和绿色校园建设。

Yuan and colleagues (2024), in a study published in the International Journal of Chinese Education, examine Beijing’s Green School Program through a study of 98 primary and secondary schools, documenting the program’s role as an ESD tool for SDG achievement. Zou and colleagues (2024), writing in the Journal of Cleaner Production, propose a four-dimensional framework — green education, green research, green campus, and green life — for the digitalization of green university initiatives, arguing that digital technologies can facilitate community engagement, support green innovation, reduce campus carbon footprints, and cultivate sustainability awareness. 袁等人（2024）在《国际华文教育期刊》发表的一项研究中，以98所中小学为研究对象，探究了北京绿色学校项目，证实该项目可作为助力可持续发展目标实现的可持续发展教育工具。邹等人（2024）在《清洁生产杂志》发文，为高校绿色建设行动的数字化构建了四维框架，涵盖绿色教育、绿色科研、绿色校园与绿色生活四个维度，并指出数字技术能够推动社会参与、助力绿色创新、降低校园碳足迹以及培养可持续发展意识。

These initiatives are real and valuable, but they do not directly address the environmental cost of the digital infrastructure itself. The tension between digital expansion and environmental sustainability is acknowledged in Chinese policy discourse but not yet resolved in practice. Over 200 Chinese universities have implemented Campus Energy Management Systems, reflecting a growing institutional awareness of energy consumption — but these systems typically do not include the energy consumed by cloud-based educational platforms, which is generated in data centres that may be located thousands of kilometres from the campus.

这些举措切实可行且具有实际价值，但并未直接解决数字基础设施自身产生的环境代价。中国政策论述中已意识到数字扩张与环境可持续发展之间存在矛盾，但在实际落地层面尚未得到解决。中国已有200多所高校部署了校园能源管理系统，体现出院校机构对能耗问题的关注度正不断提升——但这类系统通常并未纳入云端教育平台所消耗的能源，而这类平台的能耗产生于数据中心，部分数据中心甚至远在千里之外的异地。

'5. The AI Energy Paradox' 5.人工智能能源悖论

The most acute tension in green digital education is what we term the „AI energy paradox.“ Artificial intelligence is simultaneously the most energy-intensive component of digital education and the technology most frequently invoked as a solution to environmental challenges. AI systems promise to optimize energy consumption, model climate change, personalize sustainability education, and identify patterns in environmental data that human analysis cannot detect. Yet the energy required to train and operate these systems is growing at a rate that threatens to overwhelm the efficiency gains they produce. 绿色数字化教育中最尖锐的矛盾，正是我们所称的“人工智能能源悖论”。人工智能既是数字化教育里能耗最高的组成部分，又是人们最常推崇、用以解决环境难题的技术。人工智能系统有望优化能源消耗、模拟气候变化、实现可持续发展教育的个性化教学，还能挖掘出人工分析无法识别的环境数据规律。然而，训练和运行这类系统所需消耗的能源正持续激增，其增速甚至可能抵消掉系统本身带来的能效提升效益。

This paradox manifests in a form recognized in economics as a „rebound effect“ or Jevons Paradox: efficiency improvements lead to increased consumption rather than reduced resource use. A 2025 study in Frontiers in Energy Research systematically reviews 150 articles on the rebound effect in AI-driven sustainable development, finding that AI-driven efficiency reduces energy per unit of output but often leads to higher overall consumption, potentially negating environmental benefits. 这种悖论在经济学中体现为“回弹效应”（也叫杰文斯悖论）：效率提升非但没有减少资源消耗，反而会导致消耗量增加。《能源研究前沿》2025年的一项研究系统梳理了150篇关于人工智能驱动可持续发展中回弹效应的相关文献，研究发现，人工智能带来的效率提升虽然降低了单位产出的能耗，却往往会推高整体能源消耗，甚至有可能抵消其原本带来的环境效益。

For universities, the paradox is immediate. Deploying an AI-powered adaptive learning system may improve educational outcomes (as documented in the companion chapters on AI in language learning and the university of the future), but it also increases the university’s digital energy consumption. The European Commission’s sustainability frameworks and China‘s ecological civilization education both lack mechanisms for weighing these trade-offs: the environmental costs of educational technology are simply not part of the calculation. 对高校而言，这一矛盾立现。部署人工智能驱动的自适应学习系统，或许能提升教学成效（有关人工智能在语言学习及未来高校发展的配套章节已有佐证），但同时也会增加高校的数字化能耗。欧盟委员会的可持续发展框架与我国的生态文明教育体系，均缺乏权衡此类利弊取舍的相关机制：教育技术产生的环境成本，压根未被纳入考量范畴。

Selwyn (2021), in what remains the most direct academic treatment of this issue, proposes an „Ed-Tech Within Limits“ approach that requires fundamental shifts in thinking about educational technology. Rather than asking how technology can enhance education, Selwyn argues, we should ask what level of technology is compatible with environmental sustainability — and accept that the answer may involve less, rather than more, digital infrastructure. 塞尔温（2021年）在迄今为止对该问题最为直接的学术论述中，提出了一种“有限度教育科技”理念，该理念要求人们从根本上转变对教育技术的认知思路。塞尔温认为，我们不应一味探究科技如何赋能教育，而应思考何种程度的科技应用符合环境可持续发展要求，同时也要接受一个现实：解决方案或许需要“缩减而非扩充”数字基础设施规模。

'5.1 Green AI as Partial Response' 5.1 绿色人工智能作为部分解决方案

The emerging field of „Green AI“ offers technical approaches to reducing the environmental cost of artificial intelligence, though it cannot eliminate the paradox entirely. Tabbakh and colleagues (2024), in a comprehensive framework published in Discover Sustainability, review techniques including model pruning, quantization, and knowledge distillation that can substantially reduce the energy consumption of AI inference — the ongoing operational cost of running trained models. They also review tools such as CodeCarbon and Carbontracker that enable researchers to measure and report the carbon footprint of their AI experiments. 新兴的“绿色人工智能”领域为降低人工智能的环境成本提供了技术路径，但无法从根本上彻底消解这一矛盾。塔巴克及其团队于2024年在《可持续发展探索》期刊发布了一项综合性研究框架，梳理了模型剪枝、量化、知识蒸馏等技术，这些技术能够大幅降低人工智能推理阶段的能耗，也就是运行已训练模型所产生的持续运营成本。他们还综述了CodeCarbon、Carbontracker等工具，科研人员可借助这类工具测算并上报人工智能实验产生的碳足迹。

Paula and colleagues (2025), in a comparative analysis published in Scientific Reports, demonstrate that applying model compression techniques to transformer-based models can achieve a 32 percent reduction in energy consumption for models like BERT. Compressed large models can match or approach the efficiency of purpose-built small models, suggesting that educational AI applications need not rely on the most resource-intensive architectures. For universities deploying AI tutoring systems or automated assessment tools, these findings indicate that choosing efficient model architectures — or insisting that vendors demonstrate the energy efficiency of their products — could meaningfully reduce the environmental footprint of AI-powered education. 保拉及其研究团队（2025年）在《科学报告》发表的一项对比分析研究中证实，对基于Transformer架构的模型应用模型压缩技术，可使BERT等模型的能耗降低32%。经过压缩的大模型能够达到或接近专用小型模型的运行效率，这表明教育类人工智能应用无需依赖资源消耗最高的模型架构。对于部署人工智能辅导系统或自动测评工具的高校而言，该研究结果意味着，选用高效的模型架构，或是要求厂商出具产品能效相关证明，能够切实减少人工智能赋能教育所产生的环境碳足迹。

However, Green AI techniques address only the efficiency of individual systems, not the aggregate growth in AI deployment. If every AI application becomes 32 percent more efficient but the number of AI applications doubles, total energy consumption still increases — a textbook illustration of the Jevons Paradox. The technical solutions of Green AI are necessary but not sufficient; they must be combined with the institutional discipline of digital sobriety. 然而，绿色人工智能技术仅能提升单个系统的能效，却无法遏制人工智能应用规模的整体扩张。即便每款人工智能应用的能效提升32%，但如果应用数量翻倍，整体能耗依旧会上升——这正是杰文斯悖论的典型例证。绿色人工智能的技术解决方案虽必不可少，但并不充分；必须辅以数字适度理念下的制度约束才能奏效。

'6. Comparative Analysis'

6.对比分析

The EU and Chinese approaches to green digital education reflect their broader governance philosophies and reveal complementary strengths and weaknesses that merit systematic comparison. 欧盟与中国在绿色数字化教育领域的发展思路，折射出双方整体的治理理念，同时也展现出各具优势与短板的互补特质，值得进行系统性对比研究。

'6.1 Governance and Implementation'

6.1治理与实施

The EU’s framework-based approach — GreenComp, DigComp 2.2, the Green Deal, the GreenSCENT project, the EUA Green Deal Roadmap — provides conceptual clarity and citizen empowerment but struggles with implementation. Digital sobriety is recognized as a concept but not yet integrated into educational practice at scale. The environmental costs of EdTech platforms are acknowledged in academic literature but not in policy frameworks or procurement decisions. The EU approach is bottom-up: it empowers institutions and individuals to make sustainable choices but cannot compel them to do so. 欧盟基于框架的相关举措——绿色能力框架、数字能力框架2.2版、欧洲绿色协议、绿色SCENT项目、欧洲大学协会绿色协议路线图——具备清晰的理念内涵，也能赋能普通民众，但在落地实施层面面临诸多困境。数字适度理念虽已得到认可，却尚未大规模融入教育实践场景。教育科技平台产生的环境成本，虽在学术文献中有所提及，却未纳入政策框架与采购决策考量范畴。欧盟采用自下而上的推进模式：该模式赋予各类机构与个人自主做出可持续发展选择的权利，却无法强制要求其践行相关选择。

China‘s state-led approach achieves rapid deployment of both digital infrastructure and ecological civilization education, but the two streams operate largely in parallel. The National Smart Education Platform and the Green School Program coexist without a framework for addressing their potential contradictions. The 2022 Ministry of Education dual carbon work program demonstrates the capacity for rapid, system-wide curriculum reform — 21 new undergraduate programs created in a single policy cycle — but it focuses on training students for the green economy, not on reducing the environmental footprint of the educational system itself. The emphasis on ecological civilization as a comprehensive worldview provides a philosophical resource that European frameworks lack — the language of civilizational transformation — but its top-down implementation limits bottom-up innovation and critical engagement. 中国由政府主导的发展模式能够快速推进数字基础设施与生态文明教育的落地普及，但这两大发展脉络基本处于并行推进、互不交融的状态。国家智慧教育平台与绿色学校计划并行存在，却缺乏一套统筹框架来化解二者之间潜在的矛盾冲突。教育部2022年双碳工作规划彰显了全国范围内课程体系快速改革的能力，仅一个政策周期就增设21个本科专业，但该规划侧重为绿色经济培养人才，并未聚焦降低教育体系自身的环境消耗。将生态文明定位为整体性世界观，形成了欧洲相关框架所不具备的思想理论支撑，即文明转型的话语体系；不过自上而下的推行模式，也制约了自下而上的创新活力与批判性参与空间。

'6.2 The Sustainability Paradox' 6.2可持续发展悖论

A 2026 study in Humanities and Social Sciences Communications identifies a „sustainability paradox“ in digital education: environmentally, digital education can reduce travel and material impacts but increases energy demand; socially, it can widen access but deepen inequalities. This two-dimensional paradox is present in both European and Chinese contexts, though manifested differently in each. 《人文与社会科学通讯》2026年的一项研究指出了数字教育中存在的「可持续发展悖论」：从环境层面来看，数字教育能够减少出行与物资消耗带来的影响，却会增加能源消耗需求；从社会层面来看，它可以拓宽教育获取渠道，同时也会加剧社会不平等。这一二维悖论在欧洲和中国的现实场景中均有体现，只是两地的表现形式有所差异。

In the EU, the paradox manifests primarily as a tension between green aspirations and market realities. European universities increasingly adopt sustainability strategies, but their EdTech procurement decisions are driven by functionality and cost rather than environmental impact. The EUA survey found broad commitment to sustainability in principle, but specific mechanisms for reducing digital environmental footprints — energy-efficient procurement criteria, carbon budgets for cloud services, institutional policies on AI deployment — remain rare. 在欧盟，这一矛盾主要体现为绿色发展愿景与市场现实之间的冲突。欧洲高校正逐步推行可持续发展战略，但在教育科技采购决策中，往往优先考量功能与成本，而非环境影响。欧洲大学协会的调查发现，各方在理念层面普遍认同可持续发展，然而能够减少数字环境足迹的具体落地机制——如节能采购标准、云服务碳预算、人工智能应用校内管理制度等——仍十分匮乏。

In China, the paradox manifests as a tension between the state’s simultaneous commitments to digital expansion and ecological civilization. The ambition to build the world’s most digitally advanced education system is in direct tension with the ambition to achieve carbon neutrality by 2060. Wang and colleagues’ (2023) finding that electricity accounts for 77 percent of campus carbon emissions underscores the scale of this challenge: as digital infrastructure expands, so does the electricity demand that drives campus emissions. 在中国，这一悖论体现为国家在推进数字化扩张与建设生态文明两大目标之间的矛盾张力。打造全球数字化程度最高的教育体系的愿景，与2060年实现碳中和的目标形成了直接冲突。王等人（2023年）的研究发现，电力消耗占校园碳排放总量的77%，这一结论凸显了挑战的严峻程度：随着数字基础设施不断扩容，催生校园碳排放的电力需求也随之攀升。

'6.3 Research Trajectories' 6.3研究发展方向

The research landscapes in both regions reflect these governance differences. Tian and colleagues’ (2024) bibliometric analysis of Chinese ESD research reveals a field increasingly shaped by domestic political frameworks rather than international sustainability discourse. European research, by contrast, remains more internationally connected but less politically integrated — producing sophisticated analyses that may not translate into policy change. Neither research tradition has yet produced a comprehensive framework for integrating digital sobriety with broader sustainability goals in higher education. 两个地区的研究格局均体现出这类治理差异。田等人（2024）对中国可持续发展教育研究开展的文献计量分析表明，该领域越来越受国内政治框架的塑造，而非国际可持续发展话语的影响。相比之下，欧洲相关研究仍保持更强的国际关联性，但政治整合度较低，虽产出了严谨深入的分析成果，却难以转化为政策变革。两种研究范式目前均未构建出一套完整框架，用以将数字简约理念与高等教育领域更广泛的可持续发展目标相融合。

'7. Recommendations' 7.建议

Based on our comparative analysis, we propose seven recommendations for universities seeking to integrate environmental sustainability into their digital education strategies. 基于我们的对比分析，我们为希望将环境可持续性融入数字化教育战略的高校提出七项建议。

First, include digital infrastructure in institutional carbon accounting. The environmental cost of cloud computing, AI services, and platform subscriptions should be calculated and reported alongside traditional energy consumption metrics. The THE Impact Rankings and similar assessment frameworks should develop specific indicators for digital infrastructure emissions. Urbano and colleagues’ (2025) finding that 1,963 institutions now participate in impact rankings demonstrates the institutional willingness to engage with sustainability metrics — but the metrics themselves must be expanded to include the digital dimension. 首先，将数字基础设施纳入机构碳核算范畴。云计算、人工智能服务及平台订阅所产生的环境成本，应与传统能耗指标一同进行核算与上报。泰晤士高等教育影响力排名及同类评估框架，需针对数字基础设施碳排放制定专属指标。乌尔巴诺及其团队（2025年）的研究发现，目前已有1963所机构参与影响力排名，这体现了各类机构对接可持续发展指标的主观意愿，但现有指标体系必须进一步拓展，纳入数字化维度相关内容。

Second, adopt digital sobriety as a design principle for educational technology. Procurement decisions should include environmental impact assessments alongside functionality and cost. Unnecessary digital consumption — mandatory video-on policies during lectures, excessive cloud storage allocation, redundant platform subscriptions, and the routine deployment of AI tools for tasks that do not require them — should be identified and reduced. The Shift Project’s original recommendation to „buy the least powerful equipment possible and change devices as rarely as possible“ applies directly to educational technology procurement. 其次，将数字极简理念纳入教育技术的设计原则。设备采购决策除考量功能与成本外，还需纳入环境影响评估。要甄别并减少非必要的数字资源消耗，例如课堂强制开启视频、过度分配云存储空间、重复订阅各类平台、对无需使用人工智能工具的任务也常规性部署AI应用等行为。转型项目组织最初提出的建议——“购置性能够用即可的设备，尽量减少设备更换频次”，同样完全适用于教育技术的采购工作。

Third, integrate environmental awareness into digital literacy education. The environmental costs of digital activities should be part of the digital competence curriculum, not a separate sustainability module. Calis and colleagues’ (2025) finding that pre-service teachers have only moderate awareness of digital carbon footprints suggests that teacher training programs urgently need to incorporate this dimension. Students who learn about AI should also learn about AI’s energy and water consumption; students who use cloud-based learning platforms should understand the infrastructure that makes them possible. 第三，将环保意识融入数字素养教育。数字活动产生的环境成本应纳入数字能力课程体系，而非单独设立可持续发展专题模块。卡利斯等人（2025年）的研究发现，职前教师对数字碳足迹的认知仅处于中等水平，这表明教师培养项目亟需纳入这一维度的内容。学习人工智能知识的学生，也应当了解人工智能的能源与水资源消耗情况；使用云端学习平台的学生，需要知晓支撑平台运行的底层基础设施。

Fourth, develop institutional metrics for the AI energy paradox. Universities deploying AI in education should be required to demonstrate that the educational benefits justify the environmental costs — or at minimum, that the environmental costs have been calculated and minimized. Green AI techniques such as model compression (Paula et al. 2025) and efficient architectures (Tabbakh et al. 2024) should be criteria in AI procurement decisions, not afterthoughts. 第四，制定人工智能能源悖论的制度性衡量标准。在教育领域部署人工智能的高校，应当被要求证明其教育收益足以抵消环境成本，或至少已对环境成本进行核算并降至最低。模型压缩（保拉等人，2025）、高效架构设计（塔巴克等人，2024）等绿色人工智能技术，应成为人工智能采购决策的核心考量标准，而非事后补充的备选方案。

Fifth, address e-waste through institutional lifecycle management. The growing hardware intensity of digital education generates e-waste that is typically invisible in sustainability assessments. Valai Ganesh and colleagues’ (2025) demonstration that on-site recycling can achieve 90 percent material recovery rates suggests that universities could significantly reduce their e-waste footprint with relatively modest institutional investment. Extending device lifecycles through repair programs and choosing durable, upgradeable hardware would further reduce environmental impact. 第五，通过制度化的全生命周期管理处理电子废弃物。数字教育硬件使用密度不断攀升，由此产生的电子废弃物往往在可持续性评估中被忽视。瓦莱·甘内什及其团队（2025年）的研究证实，现场回收可实现90%的物料回收率，这表明高校只需投入相对有限的机构资金，就能大幅减少自身的电子废弃物排放规模。通过维修服务延长设备使用周期、选用耐用且可升级的硬件设备，还能进一步降低对环境的影响。

Sixth, support research on sustainable educational technology. The academic community should invest in research on low-energy learning technologies, efficient AI architectures for educational applications, and pedagogical approaches that achieve equivalent outcomes with less digital infrastructure. Wang and colleagues’ (2023) LEAP-LCA methodology for campus carbon assessment could be adapted to include digital infrastructure emissions, providing universities with a comprehensive tool for environmental accounting. 第六，支持可持续教育技术的研究。学术界应加大对低能耗学习技术、适用于教育场景的高效人工智能架构，以及能在更少数字基础设施支持下实现同等教学效果的教学方法的研究投入。王及其同事（2023）提出的LEAP-LCA校园碳评估方法可加以改进，纳入数字基础设施排放数据，从而为高校提供一套全面的环境核算工具。

Seventh, create EU-China dialogue on green digital education. The EU’s conceptual frameworks (GreenComp, digital sobriety) and China’s implementation capacity (dual carbon mandates, rapid curriculum reform) are complementary strengths. A structured dialogue — potentially within the framework of the Jean Monnet Centre of Excellence — could accelerate the development of practical approaches to the environmental challenges that both systems face. China’s experience with mandatory curriculum reform for the dual carbon goals could inform European efforts to scale sustainability education, while the EU’s digital sobriety framework could help China address the environmental costs of its digital education expansion. 第七，建立欧中绿色数字教育对话机制。欧盟的理念框架（绿色能力框架、数字适度理念）与中国的落地实施能力（双碳目标要求、课程体系快速改革）形成优势互补。可依托让·莫内卓越中心框架搭建常态化对话机制，助力双方加快探索应对各自教育体系所面临环境挑战的实操路径。中国围绕双碳目标推进强制性课程改革的经验，可为欧洲规模化开展可持续发展教育提供借鉴；而欧盟的数字适度理念框架，也能助力中国化解数字教育扩张带来的环境成本问题。

'8. Conclusion' 8.结论

The digital transformation of higher education is not environmentally neutral. Data centres consume 415 TWh of electricity annually and growing. AI training generates hundreds of metric tons of CO2 and evaporates hundreds of thousands of litres of freshwater. Digital content consumption accounts for 3–4 percent of per capita emissions. E-waste from educational technology is projected to grow by 50 percent within a decade. These facts are not arguments against digital education — the benefits documented in this anthology are real and significant. But they are arguments for environmental honesty: for acknowledging the costs alongside the benefits, and for designing educational technology systems that minimize environmental harm rather than ignoring it. 高等教育的数字化转型并非对环境毫无影响。数据中心每年消耗415太瓦时电力，且耗电量仍在持续增长。人工智能训练会产生数百公吨二氧化碳，消耗数十万升淡水资源。数字内容消费占人均碳排放总量的3%至4%。教育科技产生的电子废弃物预计十年内将增长50%。这些事实并非要否定数字化教育——本文集所记录的数字化教育带来的益处真实且意义重大。但这些事实呼吁我们秉持环境层面的坦诚：在认可益处的同时正视其环境代价，并设计能够最大限度降低环境危害、而非漠视环境影响的教育科技体系。

The EU and China bring different resources to this challenge. The EU has developed sophisticated conceptual frameworks — GreenComp, DigComp 2.2, digital sobriety — that provide a language for discussing the environmental costs of digital education, and its Green Deal integration into education through projects like GreenSCENT represents a genuine, if still modest, step toward practice. China has demonstrated the capacity for rapid, system-wide curriculum reform through its dual carbon education mandate and has embedded ecological civilization in its constitutional and educational frameworks — providing a depth of political commitment that European voluntary approaches cannot match. 欧盟与中国在应对这一挑战时拥有不同的资源禀赋。欧盟已构建起完善成熟的理念框架，包括绿色素养框架、数字素养框架2.2版、数字适度理念，为探讨数字教育的环境成本提供了统一话语体系；欧盟还通过绿色可持续教育网络等项目将欧洲绿色协议融入教育领域，尽管规模尚有限，但已是迈向实践层面的实质性举措。中国依托双碳教育相关部署，展现出开展全国性课程体系快速改革的能力，并将生态文明建设纳入宪法与教育顶层框架，其背后深层次的政策推行力度，是欧洲自愿式推进模式难以企及的。

Yet neither system has developed an adequate response to the AI energy paradox — the uncomfortable reality that the most powerful educational technologies are also the most environmentally costly. Green AI techniques offer partial mitigation, but the Jevons Paradox suggests that efficiency gains will be consumed by growing demand unless institutional discipline constrains deployment. The pre-service teachers surveyed by Calis and colleagues (2025) — tomorrow’s educators — have only moderate awareness of digital carbon footprints, suggesting that the problem will persist without deliberate curricular intervention. 然而，这两种体系都未能针对“人工智能能源悖论”制定出完善的应对方案——这一令人棘手的现实是：最先进的教育技术，同时也伴随着最高的环境成本。绿色人工智能技术虽能起到部分缓解作用，但杰文斯悖论表明：若缺乏制度规范对应用规模加以约束，效率提升带来的节约会被持续增长的需求所抵消。卡利斯及其团队在2025年调研的职前教师群体（未来的教育从业者），对数字碳足迹的认知仅处于中等水平，这意味着若不通过课程设置进行主动干预，该问题还将长期存在。

Developing an adequate response to these challenges is among the most important tasks facing higher education in the coming decade. It requires not less technology but smarter technology — and the institutional willingness to ask, before every digital deployment, whether the educational benefit justifies the environmental cost. The EU-China comparison suggests that the answer will require both conceptual sophistication and implementation capacity — the strengths that each system can contribute to a shared global challenge. 未来十年，高等教育面临的最重要任务之一，就是针对这些挑战制定出完善的应对方案。这并非意味着要减少技术应用，而是要采用更智慧的技术手段——同时需要各大院校具备制度层面的自觉：在每一项数字化举措落地前，都审慎考量其教育价值是否足以抵消环境成本。从欧盟与中国的对比可以看出，想要得出合理答案，既需要先进的理念认知，也需要强大的落地执行能力，而这正是两大教育体系能够为应对全球性共同挑战各自贡献的优势所在。

'Acknowledgments' 致谢

This research was conducted within the framework of the Jean Monnet Centre of Excellence „EUSC-DEC“ (EU Grant 101126782, 2023–2026). The author thanks the members of Research Groups 1 and 5 for their contributions to the comparative analysis of sustainability and digital education policy. 本研究在让·莫内卓越中心“EUSC-DEC”项目框架下开展（欧盟资助编号：101126782，执行时间2023–2026年）。作者感谢第一研究小组与第五研究小组全体成员，为可持续发展与数字化教育政策的比较分析工作所作出的贡献。

'References' 参考文献

Bianchi, G., Pisiotis, U. & Cabrera Giraldez, M. (2022). GreenComp: The European sustainability competence framework. Publications Office of the European Union, Luxembourg. JRC128040. DOI: 10.2760/13286

比安基(G.)、皮西奥蒂斯(U.)、卡夫雷拉·希拉尔德斯(M.)（2022）。绿色素养框架：欧洲可持续发展能力框架。欧盟出版办公室，卢森堡。欧盟联合研究中心文献编号：JRC128040。数字对象标识符：10.2760/13286。

Calis, S., Kahraman, N., Zeren Ozer, D. & Ergul, N. R. (2025). Determining the digital carbon footprint awareness of pre-service teachers. Humanities and Social Sciences Communications, 12, Article 1678. DOI: 10.1057/s41599-025-05944-z 卡利斯(S.)、卡拉曼(N.)、泽伦·厄泽尔(D.)、埃尔古尔(N. R.)（2025）。职前教师数字碳足迹认知水平研究。《人文与社会科学通讯》，第12卷，文章编号1678。数字对象标识符：10.1057/s41599-025-05944-z。

Caird, S. & Lane, A. (2024). Digital learning, face-to-face learning and climate change. Future Healthcare Journal, 11(3), 100156. DOI: 10.1016/j.fhj.2024.100156 凯尔德、莱恩.（2024）.数字化学习、面对面学习与气候变化.未来医疗期刊,11(3),100156.数字对象标识符:10.1016/j.fhj.2024.100156。

De Vries, A. (2026). The carbon and water footprints of data centers and what this could mean for artificial intelligence. Patterns, 7(1), 101430. DOI: 10.1016/j.patter.2025.101430

德弗里斯·A.（2026）。数据中心的碳足迹与水足迹及其对人工智能的潜在影响。模式期刊，7卷第1期，101430。数字对象标识符：10.1016/j.patter.2025.101430。

European Commission. (2021–2024). GreenSCENT — Smart Citizen Education for a Green Future. Horizon 2020 Project ID: 101036480. 欧盟委员会。（2021–2024）。绿色智慧公民教育助力绿色未来（GreenSCENT）。地平线2020计划 项目编号：101036480。

European University Association. (2023). A Green Deal Roadmap for Universities. Brussels: EUA. 欧洲大学协会。（2023）。《高校绿色转型路线图》。布鲁塞尔：欧洲大学协会。

IEA. (2025). Energy and AI. IEA, Paris. Available at: https://www.iea.org/reports/energy-and-ai 国际能源署. (2025). 能源与人工智能. 国际能源署，巴黎. 获取自：https://www.iea.org/reports/energy-and-ai。

Istrate, R. et al. (2024). The environmental sustainability of digital content consumption. Nature Communications, 15, Article 3724. DOI: 10.1038/s41467-024-47621-w 伊斯特拉特（R.）等人（2024）。数字内容消费的环境可持续性。《自然-通讯》，第15卷，文章编号3724。数字对象标识符：10.1038/s41467-024-47621-w。

Le Goff, T. (2025). Not greenwashing, but still… A closer look at big tech’s 2025 sustainability reports. Internet Policy Review, https-policyreview. 勒戈夫·T（2025）。并非漂绿行为，但依旧存疑……深度审视大型科技企业2025年可持续发展报告。《互联网政策评论》，https-policyreview。

Li, P., Yang, J., Islam, M. A. & Ren, S. (2023). Making AI Less „Thirsty“: Uncovering and Addressing the Secret Water Footprint of AI Models. Communications of the ACM. DOI: 10.1145/3724499 李平、杨健、伊斯兰·M·A、任硕（2023）。让人工智能不再“耗水”：揭示并解决人工智能模型隐秘的水足迹问题。《美国计算机协会通讯》。数字对象标识符：10.1145/3724499。

McDonagh, S. A., Caforio, A. & Pollini, A. (Eds.) (2024). The European Green Deal in Education. Routledge. DOI: 10.4324/9781003492597 麦克多纳（S. A.）、卡福里奥（A.）、波利尼（A.）（主编）（2024）。《教育领域的欧洲绿色协议》。劳特利奇出版社。数字对象标识符：10.4324/9781003492597。

Ministry of Education of the People’s Republic of China. (2022). Work Program for Building a Strong Carbon Peak Carbon Neutral Higher Education Talent Training System [加强碳达峰碳中和高等教育人才培养体系建设工作方案]. Beijing: Ministry of Education. 中华人民共和国教育部（2022）加强碳达峰碳中和高等教育人才培养体系建设工作方案[加强碳达峰碳中和高等教育人才培养体系建设工作方案].；北京：中华人民共和国教育部。

NPR. (2024, July 12). AI brings soaring emissions for Google and Microsoft. NPR. 美国国家公共广播电台（2024年7月12日）人工智能导致谷歌与微软碳排放激增；美国国家公共广播电台。

Patterson, D. et al. (2021). Carbon Emissions and Large Neural Network Training. arXiv:2104.10350. DOI: 10.48550/arXiv.2104.10350 帕特森（D.）等人（2021）. 碳排放与大型神经网络训练. arXiv预印本:2104.10350. 数字对象唯一标识符: 10.48550/arXiv.2104.10350 Paula, E., Soni, J., Upadhyay, H. & Lagos, L. (2025). Comparative analysis of model compression techniques for achieving carbon efficient AI. Scientific Reports, 15, Article 23461. DOI: 10.1038/s41598-025-07821-w。

Selwyn, N. (2021). Ed-Tech Within Limits: Anticipating educational technology in times of environmental crisis. E-Learning and Digital Media, 18(5), 496–510. DOI: 10.1177/20427530211022951 塞尔温·N（2021）。有限视域下的教育科技：环境危机时代的教育技术前瞻。《电子学习与数字媒体》，18(5)，496–510。数字对象唯一标识符：10.1177/20427530211022951。

Strubell, E., Ganesh, A. & McCallum, A. (2019). Energy and Policy Considerations for Deep Learning in NLP. Proceedings of the 57th Annual Meeting of the ACL, pp. 3645–3650. DOI: 10.18653/v1/P19-1355 斯特鲁贝尔、E.，加内什、A. 与麦卡勒姆、A.（2019）。自然语言处理中深度学习的能耗与政策考量。《第57届计算语言学协会年会论文集》，第3645–3650页。数字对象唯一标识符：10.18653/v1/P19-1355。

Tabbakh, A., Al Amin, L., Islam, M., Mahmud, G. M. I., Chowdhury, I. K. & Mukta, M. S. H. (2024). Towards sustainable AI: A comprehensive framework for Green AI. Discover Sustainability, 5, Article 408. DOI: 10.1007/s43621-024-00641-4 塔巴克·A.、阿勒阿明·L.、伊斯兰·M.、马哈茂德·G.M.I.、乔杜里·I.K.、穆克塔·M.S.H.（2024）。迈向可持续人工智能：绿色人工智能综合框架。《可持续发展探索》，第5卷，文章编号408。数字对象唯一标识符：10.1007/s43621-024-00641-4

Thao, T. Q., Hanh, T. H. & Huy, N. N. (2025). Sustainable e-waste management in higher education institutions: case study of Ho Chi Minh City University of Technology. International Journal of Environmental Science and Technology, 22, 6423–6434. DOI: 10.1007/s13762-024-06012-w 陶·T.Q.、韩·T.H.、辉·N.N.（2025）。高等院校可持续电子废弃物管理：胡志明市理工大学案例研究。《国际环境科学与技术期刊》，第22卷，6423-6434页。数字对象唯一标识符：10.1007/s13762-024-06012-w

Tian, W., Ge, J., Zheng, X., Zhao, Y., Deng, T. & Yan, H. (2024). Understanding the landscape of education for sustainable development in China: A bibliometric review and trend analysis of multicluster topics (1998–2023). Humanities and Social Sciences Communications, 11, Article 1213. DOI: 10.1057/s41599-024-03713-y 田伟、葛健、郑鑫、赵阳、邓涛、严浩（2024）。中国可持续发展教育格局解析：多集群主题文献计量综述与趋势分析（1998-2023）。《人文社科通讯》，第11卷，文章编号1213。数字对象唯一标识符：10.1057/s41599-024-03713-y

The Shift Project. (2019). Lean ICT: Towards Digital Sobriety. Paris: The Shift Project.

UNESCO. (2024). UNESCO Prize for ICT in education steers digital learning for greening. 转联合国教科文组织（2024）。联合国教科文组织教育信息通信技术奖引领数字化教育绿色转型。

Urbano, V. M., Arena, M., Azzone, G. & Mayeres, M. (2025). Sustainable development in higher education: An in-depth analysis of Times Higher Education Impact Rankings. Journal of Cleaner Production, 501, 145302. DOI: 10.1016/j.jclepro.2025.145302 乌尔巴诺·V.M.、阿雷纳·M.、阿佐内·G.、梅耶斯·M.（2025）。高等教育可持续发展：泰晤士高等教育影响力排名深度解析。《清洁生产期刊》，第501卷，文章编号145302。数字对象唯一标识符：10.1016/j.jclepro.2025.145302

Valai Ganesh, S., Suresh, V., Rajakarunakaran, S. et al. (2025). Sustainable electronic waste management framework for academic institutions in India. Scientific Reports, 15, Article 40550. DOI: 10.1038/s41598-025-24278-z 瓦莱·加内什·S.、苏雷什·V.、拉贾卡鲁纳卡兰·S.等（2025）。印度高校可持续电子废弃物管理框架。《科学报告》，第15卷，文章编号40550。数字对象唯一标识符：10.1038/s41598-025-24278-z

Vuorikari, R., Kluzer, S. & Punie, Y. (2022). DigComp 2.2: The Digital Competence Framework for Citizens. EUR 31006 EN, Publications Office of the European Union. DOI: 10.2760/115376 沃里卡里·R.、克鲁泽尔·S.、普尼·Y.（2022）。数字素养框架2.2：公民数字能力框架。欧盟官方出版物办公室，EUR 31006 EN。数字对象唯一标识符：10.2760/115376

Wang, C., Parvez, A. M., Mou, J., Quan, C., Wang, J., Zheng, Y., Luo, X. & Wu, T. (2023). The status and improvement opportunities towards carbon neutrality of a university campus in China. Journal of Cleaner Production, 414, 137521. DOI: 10.1016/j.jclepro.2023.137521 王超、帕尔维兹·A.M.、牟健、全晨、王杰、郑宇、罗鑫、吴涛（2023）。中国大学校园碳中和现状与提升路径。《清洁生产期刊》，第414卷，文章编号137521。数字对象唯一标识符：10.1016/j.jclepro.2023.137521'

Wang, Y., Chen, X., Liu, F. & Gong, Q. (2025). Ecological Civilization Education in China. In: Peters, M. A. et al. (Eds.), Handbook of Ecological Civilization. Springer. DOI: 10.1007/978-981-97-8101-0_43-1 王洋、陈旭、刘芳、龚晴（2025）。中国生态文明教育。收录于：彼得斯·M.A.等（主编），《生态文明手册》。施普林格出版社。数字对象唯一标识符：10.1007/978-981-97-8101-0_43-1'

Williamson, B., Hogan, A. & Selwyn, N. (2025). Digital Emissions: Edtech Platforms and the Extended Carbon Relations of Higher Education Institutions. In: Critical EdTech Studies. Springer. DOI: 10.1007/978-3-031-88173-2_9 威廉姆森·B.、霍根·A.、塞尔温·N.（2025）。数字碳排放：教育科技平台与高等院校延伸碳关联。收录于：《教育科技批判研究》。施普林格出版社。数字对象唯一标识符：10.1007/978-3-031-88173-2_9

Xiao, T. et al. (2025). Environmental impact and net-zero pathways for sustainable artificial intelligence servers in the USA. Nature Sustainability, 8(12), 1541–1553. DOI: 10.1038/s41893-025-01681-y 肖腾等（2025）。美国可持续人工智能服务器的环境影响与净零排放路径。《自然·可持续性》，第8卷第12期，1541-1553页。数字对象唯一标识符：10.1038/s41893-025-01681-y

Yuan, X. et al. (2024). Promoting education for sustainable development through the green school program. International Journal of Chinese Education, 13(2). DOI: 10.1177/2212585X241259192 袁鑫等（2024）。通过绿色校园项目推进可持续发展教育。《国际华文教育期刊》，第13卷第2期。数字对象唯一标识符：10.1177/2212585X241259192'

Zhou, R. K. (2024). From Education for a Sustainable Development to Ecological Civilization in China: A Just Transition? Social Inclusion, 12, Article 7421. DOI: 10.17645/si.7421 周荣凯（2024）。从可持续发展教育到中国生态文明：公平转型之路？《社会包容》，第12卷，文章编号7421。数字对象唯一标识符：10.17645/si.7421

Zou, Y., Zhong, N., Chen, Z. & Zhao, W. (2024). Bridging digitalization and sustainability in universities: A Chinese green university initiative in the digital era. Journal of Cleaner Production, 467, 142924. DOI: 10.1016/j.jclepro.2024.142924 邹宇、钟宁、陈哲、赵伟（2024）。高校数字化与可持续发展融合：数字时代中国绿色大学建设实践。《清洁生产期刊》，第467卷，文章编号142924。数字对象唯一标识符：10.1016/j.jclepro.2024.142924

'Index' 索引 A A academic integrity 7, 14, 15, 71, 173, 175 学术诚信 7, 14, 15, 71, 173, 175 AI energy paradox 183, 184, 193, 197, 199 人工智能能源悖论 183, 184, 193, 197, 199 AI ethics 6, 37, 69, 75, 87, 144, 160, 166 人工智能伦理 6, 37, 69, 75, 87, 144, 160, 166 AI in education 7, 10, 14, 18, 39, 48, 58, 59, 75, 78, 79, 98, 174, 182, 197 人工智能教育应用 7, 10, 14, 18, 39, 48, 58, 59, 75, 78, 79, 98, 174, 182, 197

AI in higher education  18, 19, 20, 77, 165, 181, 182

高等教育人工智能 18, 19, 20, 77, 165, 181, 182 AI labour market 115 人工智能劳动力市场 115 AI literacy 6, 7, 9, 11, 17, 18, 37, 65, 69, 70, 76, 123, 124, 143, 147, 148, 149, 151, 152, 153, 157, 158, 159, 160, 161, 162, 176, 178, 180 人工智能素养 6, 7, 9, 11, 17, 18, 37, 65, 69, 70, 76, 123, 124, 143, 147, 148, 149, 151, 152, 153, 157, 158, 159, 160, 161, 162, 176, 178, 180 AI-assisted language learning 59, 75, 78, 79, 84, 90, 170 人工智能辅助语言学习 59, 75, 78, 79, 84, 90, 170 alternative education 115, 125, 129 另类教育 115, 125, 129 B B Brussels Effect 7, 16, 17, 19, 20, 44, 56 布鲁塞尔效应 7, 16, 17, 19, 20, 44, 56 C C carbon footprint 184, 186, 189, 191, 192, 194, 197, 199, 200 碳足迹 184, 186, 189, 191, 192, 194, 197, 199, 200 ChatGPT 36, 58, 61, 64, 66, 67, 70, 71, 76, 77, 79, 80, 81, 99, 100, 106, 108, 112, 172, 173, 176 ChatGPT 36, 58, 61, 64, 66, 67, 70, 71, 76, 77, 79, 80, 81, 99, 100, 106, 108, 112, 172, 173, 176 China 1, 2, 3, 4, 6, 7, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 38, 39, 42, 43, 44, 46, 48, 49, 50, 53, 54, 55, 56, 59, 61, 65, 70, 71, 74, 75, 76, 78, 79, 81, 104, 114, 115, 116, 117, 118, 119, 120, 121, 122, 124, 125, 127, 128, 129, 130, 133, 134,135, 136, 138, 139, 140, 141, 143, 144, 145, 146, 147, 148, 149, 150, 151, 153, 154, 155, 156, 158, 159, 160, 161, 162, 163, 165, 166, 167, 168, 169, 171, 172, 173, 175, 176, 177, 178, 179, 180, 181, 182, 183, 185, 190, 191, 192, 193, 195, 196, 198, 199, 200, 201, 202 中国 1, 2, 3, 4, 6, 7, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 38, 39, 42, 43, 44, 46, 48, 49, 50, 53, 54, 55, 56, 59, 61, 65, 70, 71, 74, 75, 76, 78, 79, 81, 104, 114, 115, 116, 117, 118, 119, 120, 121, 122, 124, 125, 127, 128, 129, 130, 133, 134,135, 136, 138, 139, 140, 141, 143, 144, 145, 146, 147, 148, 149, 150, 151, 153, 154, 155, 156, 158, 159, 160, 161, 162, 163, 165, 166, 167, 168, 169, 171, 172, 173, 175, 176, 177, 178, 179, 180, 181, 182, 183, 185, 190, 191, 192, 193, 195, 196, 198, 199, 200, 201, 202 China AI governance 6 中国人工智能治理 6 China digital education 148' 中国数字化教育 148 China education technology 134 中国教育技术 134 comparative education 90, 116, 148 比较教育 90, 116, 148 comparative study 20, 77, 78, 79, 182 比较研究 20, 77, 78, 79, 182 competency-based education 114, 115, 180 能力本位教育 114, 115, 180

complementarity thesis  79

互补理论 79 cross-border data flows 39, 52, 55 跨境数据流动 39, 52, 55 D D data centres 127, 184, 185, 186, 193 数据中心 127, 184, 185, 186, 193 DeepL 58, 60, 61, 66, 70, 76, 77 DeepL 58, 60, 61, 66, 70, 76, 77 DigComp 2.2 142, 147, 148, 149, 156, 158, 159, 162, 188, 195, 199, 201 数字素养框架2.2 142, 147, 148, 149, 156, 158, 159, 162, 188, 195, 199, 201 digital competence 30, 147, 148, 149, 153, 154, 157, 159, 160, 171, 197 数字素养 30, 147, 148, 149, 153, 154, 157, 159, 160, 171, 197

digital divide  128, 142, 148, 149, 153, 154, 158, 161

数字鸿沟 128, 142, 148, 149, 153, 154, 158, 161 digital education 54, 78, 79, 115, 126, 135, 136, 137, 138, 147, 150, 153, 165, 183, 184, 185, 187, 188, 192, 193, 195, 196, 197, 198, 199' 数字化教育 54, 78, 79, 115, 126, 135, 136, 137, 138, 147, 150, 153, 165, 183, 184, 185, 187, 188, 192, 193, 195, 196, 197, 198, 199 digital literacy 128, 140, 144, 147, 148, 150, 151, 153, 155, 156, 157, 158, 159, 160, 162, 163, 180, 197 数字素养能力 128, 140, 144, 147, 148, 150, 151, 153, 155, 156, 157, 158, 159, 160, 162, 163, 180, 197 digital natives 148 数字原住民 148 digital sobriety 183, 184, 185, 187, 188, 189, 194, 196, 197, 198, 199 数字适度发展 183, 184, 185, 187, 188, 189, 194, 196, 197, 198, 199 E E ecological civilization 183, 184, 185, 190, 191, 192, 193, 195, 196, 199 生态文明 183, 184, 185, 190, 191, 192, 193, 195, 196, 199 Edu-Metaverse 133, 134, 135, 136, 144, 146 教育元宇宙 133, 134, 135, 136, 144, 146 EU AI Act 6, 8, 11, 14, 17, 18, 48, 53, 69, 151, 158, 177, 178 欧盟人工智能法案 6, 8, 11, 14, 17, 18, 48, 53, 69, 151, 158, 177, 178 EU-China comparison 18, 39, 58, 165, 184, 199 中欧对比 18, 39, 58, 165, 184, 199 European digital skills 148 欧洲数字技能 148 European Union 2, 6, 7, 9, 19, 38, 39, 55, 76, 79, 96, 115, 116, 126, 130, 131, 134, 147, 149, 151, 161, 162, 181, 183, 185, 200, 201 欧盟 2, 6, 7, 9, 19, 38, 39, 55, 76, 79, 96, 115, 116, 126, 130, 131, 134, 147, 149, 151, 161, 162, 181, 183, 185, 200, 201 European universities 16, 40, 48, 50, 69, 119, 120, 133, 134, 167, 177, 187, 189, 195 欧洲高校 16, 40, 48, 50, 69, 119, 120, 133, 134, 167, 177, 187, 189, 195 European-Chinese comparison 115, 118 中欧比较研究 115, 118 F F foreign language education 78 外语教育 78 G G GDPR 3, 20, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 156, 170, 178, 179, 182 通用数据保护条例 3, 20, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 156, 170, 178, 179, 182 Gen Z 148 Z世代 148 generative AI policy 7, 165 生成式人工智能政策 7, 165 green digital education 184, 192, 193, 194, 198 绿色数字化教育 184, 192, 193, 194, 198 GreenComp 183, 184, 188, 190, 191, 195, 198, 199, 200 绿色能力框架 183, 184, 188, 190, 191, 195, 198, 199, 200 H H higher education 6, 7, 11, 13, 14, 15, 17, 19, 20, 38, 39, 40, 41, 47, 49, 51, 53, 55, 75, 99, 112, 116, 117, 121, 122, 133, 134, 135, 137, 138, 145, 146, 151, 167, 171, 172, 173, 174, 177, 181, 182, 183, 184, 185, 189, 191, 196, 198, 199, 201 高等教育 6, 7, 11, 13, 14, 15, 17, 19, 20, 38, 39, 40, 41, 47, 49, 51, 53, 55, 75, 99, 112, 116, 117, 121, 122, 133, 134, 135, 137, 138, 145, 146, 151, 167, 171, 172, 173, 174, 177, 181, 182, 183, 184, 185, 189, 191, 196, 198, 199, 201 human-AI interaction 78, 88 人机交互 78, 88 hybrid learning 39, 165, 166, 171, 172, 181 混合式学习 39, 165, 166, 171, 172, 181 I I immersive learning 134 沉浸式学习 134 L

 L
language education  58, 76, 78, 79, 87, 92, 93, 95, 96, 162
语言教育  58, 76, 78, 79, 87, 92, 93, 95, 96, 162
learning analytics  38, 39, 40, 46, 47, 49, 52, 53, 54, 56, 167, 178, 179
学习分析  38, 39, 40, 46, 47, 49, 52, 53, 54, 56, 167, 178, 179
lifelong learning  115, 120, 122, 129, 131 
终身学习  115, 120, 122, 129, 131
M
 M
machine translation  58, 59, 60, 63, 64, 65, 66, 67, 69, 73, 74, 75, 76
 机器翻译  58, 59, 60, 63, 64, 65, 66, 67, 69, 73, 74, 75, 76'
micro-credentials  114, 115, 116, 128, 131, 180
 微证书  114, 115, 116, 128, 131, 180
P
P
PIPL  3, 20, 38, 39, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 156, 179, 182
个人信息保护法  3, 20, 38, 39, 40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 156, 179, 182
post-editing  58, 59, 62, 64, 67, 68, 69, 70, 72, 74, 76
译后编辑  58, 59, 62, 64, 67, 68, 69, 70, 72, 74, 76
privacy  13, 15, 18, 19, 39, 41, 43, 44, 46, 47, 50, 52, 53, 55, 56, 124
隐私保护  13, 15, 18, 19, 39, 41, 43, 44, 46, 47, 50, 52, 53, 55, 56, 124
proctoring  6, 7, 8, 9, 14, 15, 20, 38, 39, 41, 47, 48, 54
 在线监考  6, 7, 8, 9, 14, 15, 20, 38, 39, 41, 47, 48, 54
S
S
sensory modalities  78
感官模态  78'
smart campus  165, 167, 168, 178, 180, 182
智慧校园  165, 167, 168, 178, 180, 182
smart classrooms  134
智慧教室  134
smart education platform  134
智慧教育平台  134'
student attitudes  78
 学生态度  78'
student data protection  39
学生数据保护  39
sustainability  27, 30, 157, 166, 183, 184, 185, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 202
可持续发展  27, 30, 157, 166, 183, 184, 185, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 202
T 
T
translation literacy  58, 65, 74 
翻译素养  58, 65, 74
U
U
university transformation  165, 166, 181
大学转型  165, 166, 181
V
V
virtual reality  22, 133, 134, 145, 146, 166, 184
虚拟现实  22, 133, 134, 145, 146, 166, 184
VR effectiveness  134
虚拟现实应用成效  134
W
W
workforce transformation  115
劳动力转型  115
X
X
XR  133, 134, 136, 137, 145
扩展现实  133, 134, 136, 137, 145