Researchers proposed generalized green and red AI concepts. They extended the taxonomy to big data and ICT. The framework covers environmental, economic, social, and technical dimensions. It uses electricity, carbon, water, and e-waste as footprint indicators. The paper analyzes risk propagation across AI, big data, and ICT stacks. It also examines service-versus-use boundary impacts. Countermeasures are organized into a four-tier taxonomy. The framework aims to enable sustainability-by-design without sacrificing utility.
What it unlocks
A unified framework to assess and prioritize sustainability interventions across AI, big data, and ICT stacks.
- 6.7–12% of U.S. electricity by 2028
- 4–6 billion m³ AI water demand by 2027
- 4.4% U.S. electricity in 2023
Sources