The Environmental Impact of ChatGPT
Summary
Discusses the carbon and water footprints associated with training and running large language models like ChatGPT, emphasizing the need for transparency and sustainable AI practices.
Key quotes
ChatGPT emits 8.4 tons of carbon dioxide per year, more than twice the amount that is emitted by an individual
Microsoft used approximately 700,000 litres of freshwater during GPT-3’s training in its data centres
The article details the energy and water requirements for AI data centers. It highlights a University of California, Riverside study regarding the water footprint of GPT models.