🤖 AI Summary
This study addresses the lack of standardized methodologies for Scope 3 Category 1 carbon accounting in enterprise AI inference services, where current practices often overestimate emissions by 10–40× due to reliance on coarse-grained industry-average emission factors. The authors propose the first four-tier hierarchical accounting framework aligned with CSRD compliance requirements, which degrades systematically—from a token-level physical energy consumption model to expenditure-driven environmentally extended input-output (EEIO) analysis—based on data availability. Integrating GPU energy benchmarks (ML.ENERGY v3), regional grid carbon intensities (EPA eGRID, Ember), and water efficiency metrics, the framework reveals a previously undocumented carbon-water trade-off: low-carbon regions may entail high water consumption, influencing data center siting decisions. An empirical assessment of a 200-person European firm shows annual AI inference emissions under 1 ton CO₂e, indicating that regulatory challenges stem from methodological gaps rather than emission magnitude, while also uncovering water-carbon synergies overlooked by mainstream ESG tools.
📝 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.