Accounting for AI Inference in Corporate GHG Inventories: A Four-Tier Methodology for Scope 3 Category 1 Reporting

• URL: http://arxiv.org/abs/2606.10660v1
• Date: Tue, 09 Jun 2026 10:08:36 GMT
• Title: Accounting for AI Inference in Corporate GHG Inventories: A Four-Tier Methodology for Scope 3 Category 1 Reporting
• Abstract summary: AI inference services fall under Scope 3 Category 1 under the Corporate Sustainability Reporting Directive (CSRD)<n>Current practice either omits the category entirely or applies a generic economic input-output factor calibrated to the ICT sector as a whole.<n>We propose a four-tier framework that matches estimation precision to the data organisations can realistically obtain.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• 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. Yet no standardised methodology exists for including them in corporate GHG inventories. Current practice either omits the category entirely or applies a generic economic input-output (EEIO) factor calibrated to the ICT sector as a whole, overestimating AI inference emissions by 10-40x relative to physically derived alternatives. We propose a four-tier framework that matches estimation precision to the data organisations can realistically obtain, progressing from direct token-based physical estimation -- using GPU energy benchmarks and regional grid carbon intensities -- down to a spend-based EEIO fallback for services where no usage data exists. Emission factors are derived from peer-reviewed GPU energy benchmarks (ML.ENERGY Leaderboard v3), confirmed grid carbon intensities (EPA eGRID 2023; Ember 2023), and published water use effectiveness data (Li et al., 2025). Applied to a 200-person European firm, the framework yields a total below 1 tCO2e, illustrating that the compliance challenge is methodological rather than magnitude-driven. We further document a water-carbon trade-off that current ESG tools do not surface: Sweden's hydro-dominated grid delivers the lowest carbon intensity in our dataset but the highest water footprint, with direct implications for data centre location strategy.

Related papers

• Toward Sustainable AI Deployment: A Carbon-Aware Decision Framework for Enterprise Supply Chain Systems [2.9152186617858304]
  We develop a Green IS design artifact that operationalizes sustainable AI governance for enterprise procurement.<n>We benchmark six large language models across supply chain tasks, simultaneously measuring decision quality and estimated generation-related operational carbon.<n>Our "benchmark first, select green" principle suggests that environmental responsibility and decision quality can be mutually reinforcing.
  arXiv  Detail & Related papers  (2026-09-14T01:03:47Z)
• Agents' Last Exam [265.9133708028224]
  This paper introduces Agents' Last Exam (ALE), a benchmark designed to evaluate AI agents on long-horizon, economically valuable, real-world tasks.<n>Developed in collaboration with 250+ industry experts, ALE covers non-physical industries defined with reference to O*NET / SOC 2018.<n>Current results show that the hardest tier remains far from saturated: across mainstream harness and backbone configurations, the average full pass rate is 2.6%.
  arXiv  Detail & Related papers  (2026-06-03T20:20:46Z)
• The Hidden Cost of Thinking: Energy Use and Environmental Impact of LMs Beyond Pretraining [59.27959962602072]
  Olmo 3 is a family of 7 billion and 32 billion parameter models in both instruction-following and reasoning variants.<n>We estimate our model development process consumed 12.3 GWh of datacenter energy, emitted 4,251 tCO2eq, and consumed 15,887 kL of water, with water consumption driven entirely by power generation infrastructure rather than data center cooling.
  arXiv  Detail & Related papers  (2026-05-01T23:24:23Z)
• Quantifying the Climate Risk of Generative AI: Region-Aware Carbon Accounting with G-TRACE and the AI Sustainability Pyramid [2.2999148299770047]
  GenAI represents a rapidly expanding digital infrastructure whose energy demand and associated CO2 emissions are emerging as a new category of climate risk.<n>This study introduces G-TRACE, a cross-modal, region-aware framework that quantifies training- and inference-related emissions.<n>We propose the AI Sustainability Pyramid, a seven-level governance model linking carbon accounting metrics with operational readiness, optimization, and stewardship.
  arXiv  Detail & Related papers  (2025-11-06T19:52:02Z)
• Environmental Impact of CI/CD Pipelines [45.02374935546107]
  GitHub Actions ecosystem results in a substantial carbon and water footprint.<n>To provide perspective, the carbon footprint in the most likely scenario is equivalent to the carbon captured by 7,615 urban trees in a year.<n>Key recommendations include deploying runners in regions whose energy production has a low environmental impact.
  arXiv  Detail & Related papers  (2025-10-30T12:02:09Z)
• A Graph-Retrieval-Augmented Generation Framework Enhances Decision-Making in the Circular Economy [40.04207519131063]
  CircuGraphRAG is a retrieval-augmented generation (RAG) framework that grounds LLMs outputs in a domain-specific knowledge graph for the circular economy.<n>This graph connects 117,380 industrial and waste entities with classification codes and GWP100 emission data, enabling structured multi-hop reasoning.
  arXiv  Detail & Related papers  (2025-06-01T07:49:47Z)
• How Hungry is AI? Benchmarking Energy, Water, and Carbon Footprint of LLM Inference [0.0]
  This paper introduces a novel infrastructure-aware benchmarking framework for quantifying the environmental footprint of AI inference across 30 state-of-the-art models as deployed in commercial data centers.<n>Our results show that o3 and DeepSeek-R1 emerge as the most energy-intensive models, consuming over 33 Wh per long prompt, more than 70 times the consumption of GPT-4.1 nano, and that Claude-3.7 Sonnet ranks highest in eco-efficiency.<n>These findings illustrate a growing paradox: Although AI is becoming cheaper and faster, its global adoption drives disproportionate resource consumption.
  arXiv  Detail & Related papers  (2025-05-14T17:47:00Z)
• Group Reasoning Emission Estimation Networks [11.479035866165926]
  We introduce an AI-driven carbon accounting framework that standardizes enterprise-level emission estimation.<n>We use a novel reasoning approach with large language models (LLMs)<n> Experiments on 1,114 NAICS categories yield state-of-the-art performance.
  arXiv  Detail & Related papers  (2025-02-08T09:02:43Z)
• Supply chain emission estimation using large language models [15.605998085195314]
  We propose a first-of-a-kind framework that uses domain-adapted NLP foundation models to estimate Scope 3 emissions. We compare the performance of the proposed framework with the state-of-the-art text classification models such as TF-IDF, word2Vec, and Zero shot learning.
  arXiv  Detail & Related papers  (2023-08-03T13:06:37Z)
• An Energy and Carbon Footprint Analysis of Distributed and Federated Learning [42.37180749113699]
  Classical and centralized Artificial Intelligence (AI) methods require moving data from producers (sensors, machines) to energy hungry data centers. Emerging alternatives to mitigate such high energy costs propose to efficiently distribute, or federate, the learning tasks across devices. This paper proposes a novel framework for the analysis of energy and carbon footprints in distributed and federated learning.
  arXiv  Detail & Related papers  (2022-06-21T13:28:49Z)
• Measuring the Carbon Intensity of AI in Cloud Instances [91.28501520271972]
  We provide a framework for measuring software carbon intensity, and propose to measure operational carbon emissions. We evaluate a suite of approaches for reducing emissions on the Microsoft Azure cloud compute platform.
  arXiv  Detail & Related papers  (2022-06-10T17:04:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.