# Carbon Reporting Practices in the NHS: Emissions and Omissions Relating to Artificial Intelligence

## TL;DR

This paper examines how AI's carbon footprint is overlooked in NHS sustainability reports and suggests ways to include it for better climate accountability.

## Contribution

The paper identifies three key gaps in NHS carbon reporting related to AI and proposes practical measures to address them.

## Key findings

- NHS carbon reporting lacks granularity, hiding the high energy use of specific AI workloads.
- Upstream emissions from AI hardware like GPUs are often excluded unless owned by the trust.
- Unprocured generative AI tools like ChatGPT may emit ≈ 349t CO₂e annually in primary care alone.

## Abstract

Artificial intelligence (AI) is being rolled out across the UK National Health Service (NHS) to improve efficiency; yet, its carbon footprint is largely invisible within mandatory Green Plan reporting. This work shows where NHS carbon reporting omits AI-related emissions and proposes feasible accounting and procurement measures that allow trusts to assess whether AI adoption advances or undermines net zero. A review of NHS sustainability guidance, the Department for Environment, Food & Rural Affairs conversion factors, and recent evidence on AI energy use shows that current Scopes 1-3 accounting omits substantial emissions at 3 points. First, a lack of granularity provides averages that can obscure the extreme energy intensity of certain AI workloads. Second, life-cycle emissions from specialized hardware (eg, graphics processing units) are often excluded unless trusts own the equipment, ignoring upstream manufacturing impacts. Third, widespread use of unprocured generative AI tools is unmeasured; extrapolating general practice survey data suggests that ChatGPT queries alone could release ≈ 349t CO₂e per year in primary care. To close these gaps, we propose three potential ways to help reduce these reporting gaps: (1) AI-specific carbon disclosure clauses in vendor contracts, (2) inclusion of cradle-to-grave emission factors for AI hardware in Scope 3 reporting, and (3) lightweight monitoring of external AI traffic (while recognizing potential ethical issues with this). Implementing these measures would give health care leaders a more accurate baseline against which to judge whether AI supports or undermines the NHS net-zero target.