EN CZ
Sweet

Tackling Climate Change with Machine Learning: A Comprehensive Roadmap

Rolnick et al. (2022) assembled a comprehensive survey with contributions from leading AI and climate researchers, identifying 13 key areas where machine learning can meaningfully contribute to climate change mitigation and adaptation:

  • Electricity systems: AI-optimized power grids can reduce energy waste by 10-15%
  • Transportation: Route optimization, autonomous vehicles, and demand prediction
  • Buildings: Smart HVAC systems that reduce energy consumption by 20-30%
  • Industry: Process optimization, supply chain efficiency, materials science
  • Agriculture: Precision farming, crop yield prediction, reducing food waste
  • Forests & Land Use: Satellite-based deforestation monitoring, carbon stock estimation
  • Climate prediction: Improved weather and climate models

The Net Question

While AI consumes significant energy (see Strubell 2019, Patterson 2021), this paper argues the net impact of well-applied AI on climate is overwhelmingly positive — the energy savings enabled by AI far exceed its own consumption.

Source

Rolnick, D. et al. (2022). ACM Computing Surveys, 55(2), 1-96.

Connected Research

You may also like