REVIEW 10 cited by
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Measuring the environmental impact of delivering AI at Google Scale
read the original abstract
The transformative power of AI is undeniable - but as user adoption accelerates, so does the need to understand and mitigate the environmental impact of AI serving. However, no studies have measured AI serving environmental metrics in a production environment. This paper addresses this gap by proposing and executing a comprehensive methodology for measuring the energy usage, carbon emissions, and water consumption of AI inference workloads in a large-scale, AI production environment. Our approach accounts for the full stack of AI serving infrastructure - including active AI accelerator power, host system energy, idle machine capacity, and data center energy overhead. Through detailed instrumentation of Google's AI infrastructure for serving the Gemini AI assistant, we find the median Gemini Apps text prompt consumes 0.24 Wh of energy - a figure substantially lower than many public estimates. We also show that Google's software efficiency efforts and clean energy procurement have driven a 33x reduction in energy consumption and a 44x reduction in carbon footprint for the median Gemini Apps text prompt over one year. We identify that the median Gemini Apps text prompt uses less energy than watching nine seconds of television (0.24 Wh) and consumes the equivalent of five drops of water (0.26 mL). While these impacts are low compared to other daily activities, reducing the environmental impact of AI serving continues to warrant important attention. Towards this objective, we propose that a comprehensive measurement of AI serving environmental metrics is critical for accurately comparing models, and to properly incentivize efficiency gains across the full AI serving stack.
Forward citations
Cited by 10 Pith papers
-
Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity
Rescaled ASGD recovers convergence to the true global objective by rescaling worker stepsizes proportional to computation times, matching the known time lower bound in the leading term under non-convex smoothness and ...
-
OpenG2G: A Simulation Platform for AI Datacenter-Grid Runtime Coordination
OpenG2G is a new extensible simulation platform that lets users implement and compare classic, optimization, and learning-based controllers for AI datacenter power flexibility coordinated with the grid.
-
Scalable LLM-based Coding of Dialogue in Healthcare Simulation: Balancing Coding Performance, Processing Time, and Environmental Impact
Larger batch sizes for LLM dialogue coding in healthcare simulations improve speed and reduce energy consumption while decreasing coding accuracy compared to human labels.
-
All LCA models are wrong. Are some of them useful? Towards open computational LCA in ICT
ICT LCAs form error-prone model systems requiring explicit dependency graphs and open versioned repositories for credible assessments.
-
Transparent Screening for LLM Inference and Training Impacts
The paper proposes a transparent proxy framework for estimating LLM inference and training environmental impacts from natural-language application descriptions.
-
Energy Scaling Laws for Diffusion Models: Quantifying Compute in Image Generation
An adapted scaling law predicts GPU energy consumption for diffusion model inference with R² > 0.9 within architectures and strong cross-architecture generalization.
-
Evaluation of ML Resource Utilization Requires Model Life Cycle Assessment
The paper calls for life cycle assessment to capture embodied hardware costs and full pipeline operational costs in AI development and deployment.
-
Greening AI Inference with Accuracy and Latency-aware User Incentives
A framework for two-tier AI service subscriptions that offer discounts for accepting lower quality or higher latency inference to reduce carbon emissions during high-intensity periods.
-
From Cradle to Cloud: A Life Cycle Review of AI's Environmental Footprint
A review of AI sustainability studies finds inconsistent life cycle definitions and predominant reliance on coarse CO2e proxies, with limited coverage of water, materials, and multi-impact assessments.
-
AgentStop: Terminating Local AI Agents Early to Save Energy in Consumer Devices
AgentStop uses execution signals to early-terminate failing local LLM agent trajectories, cutting energy use 15-20% with minimal utility loss.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.