Charles Humble
Techie, podcaster, editor, author and consultant
Charles is available to give these talks at conferences and events.
Presentations
Green AI: Making Machine Learning Environmentally Sustainable
Abstract
AI's rapid growth is having a significant and largely hidden environmental cost — Microsoft and Google have both reported carbon emissions surging by 30-50% in recent years, driven primarily by AI infrastructure. In this talk, Charles explores the environmental impact of large language models and offers practical strategies for reducing it at each stage of the AI lifecycle. Topics include demand shifting and shaping, model compression techniques (pruning, distillation, and quantisation), federated learning, speculative decoding, and how choosing where and when to run workloads can reduce software carbon emissions by up to 99%.
Slides
Resources
Data sources:
IEA Data Centres and Data Transmission Networks
Microsoft sustainability report
Google sustainability report
Hockey stick
Putting a CO2 price on computation
Tools:
Electricity Maps
WattTime
AQT
ML.Energy Leaderboard
Techniques:
Demand shifting and shaping
Federated Learning
AI Consumes Lots of Energy. Can It Ever Be Sustainable?
Speculative Decoding
Books:
The Developer's Guide to Cloud Infrastructure, Efficiency and Sustainability by Charles Humble
Kubernetes at the Edge: Container Orchestration at Scale by Charles Humble
Building Green Software by Anne Currie, Sarah Hsu, Sara Bergman
Not the End of the World by Hannah Richie
Newsletter:
AI for the rsst of us
Video
