[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"$fHzC4az6Uy8mBnBIi3ivQZKOHRZnDS0P0ABVG0lBVpe8":3},{"code":4,"msg":5,"data":6},200,"操作成功",{"id":7,"title":8,"content":9,"digest":10,"source":10,"coverPath":11,"thumbsCoverPath":12,"isTop":13,"isShow":14,"baseClick":13,"clickCount":15,"createTime":16,"typeId":17,"isNewest":18,"newsInfoTypeRespVo":19,"voiceUrl":22,"voiceSize":23,"taskId":24,"releaseTime":25,"titleEn":26,"contentEn":27,"voiceUrlEn":28,"taskIdEn":29,"voiceSizeEn":30},1267,"AI 请求平均仅耗 5 滴水？谷歌AI环境影响报告引发争议","\u003Cp>\u003Cstrong class=\"ql-lineHeight-1-75\" style=\"font-size: 18px; color: rgb(255, 153, 0);\">据CBS新闻报道，谷歌近日发布关于其Gemini人工智能模型环境影响的报告，称该模型在日常使用中能耗、水耗及碳排放均处于较低水平。然而，这份报告随即遭到多位环境研究专家的质疑，焦点集中在数据核算范围与方法的合理性上。\u003C\u002Fstrong>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">谷歌在报告中表示，其采用“全栈”方法核算，涵盖闲置设备、数据中心管理开销等多项指标。数据显示，用户每次向Gemini模型输入文本请求，平均耗水量为0.26毫升（约5滴水），能耗为0.24瓦时（低于观看9秒电视的用电量），碳排放为0.03克二氧化碳当量。谷歌还宣称，2024年5月至2025年5月间，每次请求的总排放量已降至此前的44分之一。\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cimg alt=\"undefined\" src=\"https:\u002F\u002Fimage.51xinwei.com\u002F2025\u002F08\u002F9f303e2336a44382915605a18f7a5e92\u002F5c467cbf3be14d35819453e3348cf050.webp\" width=\"undefined\" height=\"undefined\" style=\"display: block; margin: auto;\">\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">针对上述数据，美国加州大学河滨分校副教授Shaolei Ren指出，谷歌的核算仅统计了数据中心现场用于降温的水量，却忽略了为数据中心供电的发电厂在发电过程中消耗的水资源。他认为，这种“仅看表面”的统计方式，严重低估了AI运行的真实水足迹。Ren此前的研究显示，训练类似GPT-3的大模型，在美国高端数据中心耗水量可达70万升；即便是与ChatGPT进行一次20-50轮的对话，间接用水量也能达到500毫升。\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">除水耗争议外，专家还对谷歌的碳排放核算方法提出批评。谷歌采用“市场基准法”，允许企业通过扣除所购绿证来降低碳排放数据；而“位置基准法”则依据实际电网碳强度计算，更能反映当地真实环境影响。不少专家认为，“市场基准法”可能导致碳排放数据被人为压低，难以客观呈现AI运行对环境的实际影响。\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"color: rgb(187, 187, 187);\">【新闻来源】 \u003C\u002Fspan>\u003Cspan style=\"color: rgb(187, 187, 187); font-size: 16px;\">环球网科技 \u003C\u002Fspan>\u003Ca href=\"https:\u002F\u002Fwww.sohu.com\u002Fa\u002F926773504_99900743?scm=1019.20001.0.0.0&spm=smpc.csrpage.news-list.2.1755845338766QldRG2V\" rel=\"noopener noreferrer\" target=\"_blank\" style=\"color: rgb(187, 187, 187);\">https:\u002F\u002Fwww.sohu.com\u002Fa\u002F926773504_99900743?scm=1019.20001.0.0.0&spm=smpc.csrpage.news-list.2.1755845338766QldRG2V\u003C\u002Fa>\u003C\u002Fp>\u003Cp class=\"ql-align-justify\">\u003Cspan style=\"color: rgb(187, 187, 187);\">（本网转发此文章，旨在为读者提供更多的信息资讯，所涉内容不构成投资、消费建议。文章事实如有疑问，请与有关方核实，文章观点非本网观点，仅供读者参考。）\u003C\u002Fspan>\u003C\u002Fp>","","https:\u002F\u002Fimage.51xinwei.com\u002F2025\u002F08\u002F7b8b7165f64749138e51afd9e3fdee6c\u002FAI领域.jpg","https:\u002F\u002Fimage.51xinwei.com\u002F2025\u002F08\u002Fthumbs\u002F7b8b7165f64749138e51afd9e3fdee6c\u002FAI领域.jpg",0,1,222,"2025-08-22 18:11",2,false,{"id":17,"name":20,"enName":21},"芯位视野","Xinwei Vision","https:\u002F\u002Fxinwei-dev-test.oss-cn-shenzhen.aliyuncs.com\u002Fintelligent\u002Faudio%3Ad08e14d9-da06-4562-bd6c-562b8d397e2c%3A0.wav?Expires=1755861941&OSSAccessKeyId=LTAI5tNvY2RkKjZw4LLWsrPK&Signature=RsffC6k2890HwKSe%2BnXVAHnhLNs%3D",3574496,"d08e14d9-da06-4562-bd6c-562b8d397e2c","2025-08-22 18:09","AI request consumes only 5 drops of water on average? Google AI environmental impact report sparks controversy","\u003Cp>\u003Cstrong class=\"ql-lineHeight-1-75\" style=\"font-size: 18px; color: rgb(255, 153, 0);\">According to CBS News, Google recently released a report on the environmental impact of its Gemini artificial intelligence model, stating that the model has low energy consumption, water consumption, and carbon emissions during daily use. However, this report has been questioned by multiple environmental research experts, with the focus on the reasonableness of the data calculation scope and methods.\u003C\u002Fstrong>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">Google stated in the report that it uses a \"full-stack\" approach for calculation, covering multiple indicators such as idle equipment and data center management overheads. Data shows that each time a user inputs a text request to the Gemini model, the average water consumption is 0.26 milliliters (about 5 drops of water), energy consumption is 0.24 watt-hours (less than the electricity used to watch a 9-second TV program), and carbon emissions are 0.03 grams of carbon dioxide equivalent. Google also claims that the total emissions per request have been reduced to 1\u002F44th of the previous level between May 2024 and May 2025.\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cimg alt=\"undefined\" src=\"https:\u002F\u002Fimage.51xinwei.com\u002F2025\u002F08\u002F9f303e2336a44382915605a18f7a5e92\u002F5c467cbf3be14d35819453e3348cf050.webp\" width=\"undefined\" height=\"undefined\" style=\"display: block; margin: auto;\">\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">Regarding these data, Associate Professor Shaolei Ren from the University of California, Riverside pointed out that Google's calculation only counts the water used for cooling in the data center, but ignores the water resources consumed by power plants when generating electricity for the data center. He believes that this \"superficial\" statistical method seriously underestimates the real water footprint of AI operations. Ren's previous research showed that training a large model similar to GPT-3 can consume up to 700,000 liters of water in high-end data centers in the United States; even a single conversation with ChatGPT involving 20-50 rounds can indirectly consume 500 milliliters of water.\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"font-size: 18px;\" class=\"ql-lineHeight-1-75\">In addition to the controversy over water consumption, experts also criticized Google's method of calculating carbon emissions. Google uses the \"market-based approach,\" allowing companies to reduce carbon emission data by deducting purchased green certificates; while the \"location-based approach\" calculates based on the actual carbon intensity of the grid, which better reflects the real environmental impact locally. Many experts believe that the \"market-based approach\" may lead to artificially lowered carbon emission data, making it difficult to objectively present the actual environmental impact of AI operations.\u003C\u002Fspan>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cbr>\u003C\u002Fp>\u003Cp>\u003Cspan style=\"color: rgb(187, 187, 187);\">[News source] \u003C\u002Fspan>\u003Cspan style=\"color: rgb(187, 187, 187); font-size: 16px;\">Global Network Technology \u003C\u002Fspan>\u003Ca href=\"https:\u002F\u002Fwww.sohu.com\u002Fa\u002F926773504_99900743?scm=1019.20001.0.0.0&spm=smpc.csrpage.news-list.2.1755845338766QldRG2V\" rel=\"noopener noreferrer\" target=\"_blank\" style=\"color: rgb(187, 187, 187);\">https:\u002F\u002Fwww.sohu.com\u002Fa\u002F926773504_99900743?scm=1019.20001.0.0.0&spm=smpc.csrpage.news-list.2.1755845338766QldRG2V\u003C\u002Fa>\u003C\u002Fp>\u003Cp class=\"ql-align-justify\">\u003Cspan style=\"color: rgb(187, 187, 187);\">（This article is republished by this site to provide readers with more information. The content does not constitute investment or consumer advice. If there are any questions about the facts of the article, please verify with the relevant party. The views expressed in the article are not the views of this site, and are for reference only.）\u003C\u002Fspan>\u003C\u002Fp>","https:\u002F\u002Fxinwei-dev-test.oss-cn-shenzhen.aliyuncs.com\u002Fintelligent\u002Faudio%3A7130b429-7280-4f20-89c9-07cda0353b0b%3A0.wav?Expires=1774838488&OSSAccessKeyId=LTAI5tNvY2RkKjZw4LLWsrPK&Signature=EFV0zNVm2wZGNZszPj8mwkqaCvM%3D","7130b429-7280-4f20-89c9-07cda0353b0b",5714330]