1、Measuring the environmental impact of delivering AI at Google ScaleCooper Elsworth,Keguo Huang,David Patterson,Ian Schneider,Robert Sedivy,Savannah Goodman,Ben Townsend,Parthasarathy Ranganathan,JeffDean,Amin Vahdat,Ben Gomes,and James ManyikaGoogle,Mountain View,CA,USAAbstractThe transformative pow

2、er of AI is undeniablebut as user adoption accelerates,so does the need to understand andmitigate the environmental impact of AI serving.However,no studies have measured AI serving environmental metricsin a production environment.This paper addresses this gap by proposing and executing a comprehensi

3、ve methodologyfor measuring the energy usage,carbon emissions,and water consumption of AI inference workloads in a large-scale,AI production environment.Our approach accounts for the full stack of AI serving infrastructureincluding activeAI accelerator power,host system energy,idle machine capacity,

4、and data center energy overhead.Through detailedinstrumentation of Googles AI infrastructure for serving the Gemini AI assistant,we find the median Gemini Appstext prompt consumes 0.24 Wh of energya figure substantially lower than many public estimates.We also show thatGoogles software efficiency ef

5、forts and clean energy procurement have driven a 33x reduction in energy consumptionand a 44x reduction in carbon footprint for the median Gemini Apps text prompt over one year.We identify that themedian Gemini Apps text prompt uses less energy than watching nine seconds of television(0.24 Wh)and co

6、nsumesthe equivalent of five drops of water(0.26 mL).While these impacts are low compared to other daily activities,reducing the environmental impact of AI serving continues to warrant important attention.Towards this objective,wepropose that a comprehensive measurement of AI serving environmental m