Guidelines on the implementation of environmental efficiency criteria for artificial intelligence and other emerging technologies

Supplement 53 to ITU-T L-series Recommendations provides guidelines to policy-makers, technologists, innovators, environmentalists and other stakeholders from the technology industry, environmental sciences and policy arena on the topic of environmental efficiency criteria to assess the environmental impacts of artificial intelligence and other emerging technologies. These guidelines aim to serve as common factors for the above-mentioned stakeholders to consider while developing, deploying and promoting any piece of technology into the market and society, rather providing than a comprehensive list of criteria. While "emerging technologies" is a broad term, this Supplement identifies a few sample technologies through their accordant applications and areas of work in 16 applicable industry domains, which stakeholders can use as references to improve the environmental efficiency of their own technological products and/or services. When discussing environmental efficiency, this Supplement approaches environmental efficiency criteria from an adjusted model of life-cycle assessment of a product, within which three stages of environmental impacts – materials, use and end of life – are examined. The Supplement provides both long-term and short-term strategies, which include not only specific examples for certain technologies addressing the three stages of environmental efficiency, but also an instrument to be used to localize such guidelines as well as to allow global benchmarking.

https://www.itu.int/ITU-T/recommendations/rec.aspx?rec=15169

Computer processing, data management and energy perspective

Supplement 52 to ITU-T L-series Recommendations proposes a set of good practices to improve the energy efficiency of cyber-physical applications – making use of Internet of things (IoT), artificial intelligence (AI), and digital twins. First, the Supplement introduces the cyber-physical paradigm, engineering reference framework, and a couple of system deployments models. Secondly, it defines three end-to-end use case typologies to be addressed (i.e., monitoring applications using smart IoT systems and AI software; smart applications using Edge computing and cloud data centre; and simulation applications using digital twin pattern). Energy efficiency practices are discussed adopting a circular value-chain model that consists of three main steps: data storage; data transfer/move; and data processing/analytics. Finally, this Supplement offers a set of recommended practices relating to each component of the three end-to-end use case typologies.

 https://www.itu.int/ITU-T/recommendations/rec.aspx?rec=15168

Recommendation ITU-T L.1400: Overview and general principles of methodologies for assessing the environmental impact of information and communication technologies

Recommendation ITU-T L.1400 presents the general principles on assessing the environmental impact of information and communication technologies (ICTs) and outlines the different methodologies that have been developed in the L.1400-series: • Assessment of the environmental impact of ICT goods, networks and services • Assessment of the environmental impact of ICT projects • Assessment of the environmental impact of ICT in organizations • Assessment of the environmental impact of ICT in cities • Assessment of the environmental impact of the ICT sector • Assessment on how the use of ICT solutions impacts greenhouse gas (GHG) emissions of other sectors • Decarbonization trajectories for the ICT sector • Net zero guidance for ICT organizations • Guidance on how to address the ITU's Connect 20xx targets The Recommendation describes the intended usage of each Recommendation and the connections between them. Finally, it lists ongoing work items. 

https://www.itu.int/ITU-T/recommendations/rec.aspx?rec=15182

ITU-T Recommendations: Green ICT Standards and Supplements

 https://www.itu.int/net/ITU-T/lists/standards.aspx?Group=5&&&&Domain=28