The paper proposes a token‑based method for including purchased AI inference in corporate Scope 3 greenhouse gas inventories, offering a structured four‑term factor that accounts for serving electricity, embodied hardware, data‑centre construction, and disclosed training runs. It provides an open dataset with cradle‑to‑gate boundaries, four tiers of activity data, and a rule to map model identifiers and regions to specific factors, revealing up to twelve‑fold variation between model classes and nine‑fold variation between regions. Example calculations show annual emissions of 8.5 kg CO₂e for a 200‑person firm and 2.1–8.3 t CO₂e for developers using mid‑class or frontier‑class models, respectively.
By Guillermo Llopis-Garcia (SOMA AI, Barcelona)
The paper introduces a multi-objective reinforcement learning framework for planning AI data center build-outs, treating the long-term infrastructure decisions as a tiered Markov decision process. It proposes a preference-conditioned agent that can generate policies balancing cost, carbon, water, and reliability, and validates the approach on real site data across different commitment regimes. The study also presents a lightweight counterfactual rollout model to assess long-horizon dynamics and releases the associated environment, agents, and evaluation tools.
By Satyapragnya Kar, Rajarshi Das, PR Panigrahi
The paper introduces a unified framework that expands the concepts of green and red AI to include big data and ICT, proposing generalized green AI (GGAI) and generalized red AI (GRAI) along with related red categories for BD, ICT, and digital technologies. It analyzes lifecycle impacts—electricity, carbon, water, and e‑waste—across AI‑BD‑ICT stacks and service‑use boundaries, and presents a four‑tier taxonomy of countermeasures aligned with an Environmental‑Economic‑Social‑Technical/Operational framework. The framework aims to guide sustainability‑by‑design practices without compromising utility.
By Jinsong Wu, Yixin Xiao, Haowei Bi, Huamin Fang
SustainAI is a water‑aware, closed‑loop framework that embeds environmental accountability into AI deployment. It combines real‑time water metering, a hallucination‑aware penalty model, and a water‑aware routing algorithm that considers regional water stress. In tests with small language models, water footprints varied 11‑fold across data centers, and 1,335 inference runs consumed about 399 mL of water but yielded only 240 correct outputs, highlighting the resource cost of inaccurate responses.
By Farnaz Farid, Tashfia Towkee, Sania Nasreen, Sami bin Azad
The paper presents an LLM-based predictive scheduling system that forecasts execution time and energy consumption from source code, aiming to improve data center sustainability. By integrating these predictions into a real-time GPU allocation algorithm, the system reduces both energy use and queuing delays. In a collaboration with a data center, the approach achieved a 32% drop in energy consumption and a 30% reduction in waiting time.
By Hanzhao Wang, Jingxuan Wu, Yumeng Li, Yu Pan, Guanting Chen
arXiv:2608. 12915v1 Announce Type: cross Abstract: The rapid growth of LLM inference is shifting sustainability concerns from one-time training to continuous serving, where infrastructure decisions shape energy use, carbon emissions, water consumption, and service quality.
By Nicoletta Tsiopani, Moysis Symeonides, George Pallis, Marios D. Dikaiakos
arXiv:2606. 14707v1 Announce Type: cross Abstract: AI training and deployment consume substantial electricity, but carbon outcomes remain weakly integrated into routine model development decisions.
By Yuxin Chen (University of Helsinki, Finland), Hao Gao (Independent Researcher), Chujie Zou (University of Helsinki, Finland)
arXiv:2607. 05408v1 Announce Type: cross Abstract: The environmental impact of training large language models (LLMs) is increasingly scrutinised, yet most published estimates focus on operational energy and disclose little about manufacturing (embodied) emissions, water consumption, or the underlying highperformance computing (HPC) infrastructure.
By Marc L\'eobet, Pierre-Fran\c{c}ois Lavall\'ee, Jean-Pierre Lorr\'e
arXiv:2606. 10660v1 Announce Type: cross Abstract: AI inference services -- API subscriptions, enterprise chat tools, and SaaS products with embedded AI features -- fall unambiguously within Scope 3 Category 1 under the Corporate Sustainability Reporting Directive (CSRD), which requires disclosure for fiscal years starting January 2024.
By Guillermo Llopis (SOMA AI, Barcelona)
arXiv:2609.33965v2 Announce Type: replace-cross
Abstract: We describe a methodology for estimating the per-token energy cost of cloud-hosted large language model (LLM) inference, separating between i...
By Joshua Horswill, Ross Hunter, Matt Clifford, James Hall
AutoEnergyDesign is an LLM‑driven framework that optimizes building energy system designs by iteratively proposing configuration changes based on prior simulation results and automatically reverting unsuccessful changes via a git‑based mechanism. The agent combines a reasoning model with explicit memory (procedural, semantic, episodic, reflective), a constrained action interface, and a keep/revert signal, and evaluates designs using a 25‑year Net Present Cost model across eight technologies. In a case study with measured building loads, the agent (using Claude Haiku 5.5, Sonnet 5.5, and Opus 5.5) achieved an NPC within 1% of the best baseline after far fewer simulations, outperforming random search, TPE Bayesian optimization, and a $(1+1)$ evolution strategy.
By Amedeo Buonanno, Salvatore Fabozzi, Maria Valenti
arXiv:2606. 00811v1 Announce Type: cross Abstract: Data centers now account for 4.
By Dana Golden, Aruna Balasubramanian, Niranjan Balasubramanian