Related Topics

Stubble Burning
2023 NOV   24
Biomass Co-firing
2023 OCT   25
2023 AUG   23

Carbon Impact of Artificial Intelligence

2023 MAY 23

Mains   > Environment & Ecology   >   Pollution   >   Air pollution


  • As big data, machine learning, and artificial intelligence continue to gain prominence in information technology, experts are raising concerns about the environmental costs of computation — primarily data and AI’s carbon footprint and greenhouse gas emissions.


  • Artificial intelligence (AI) is an inter-disciplinary branch of science concerned with building smart machines capable of performing tasks that typically require human intelligence.
  • The term was coined in 1956 by John McCarthy, who is dubbed as the ‘Father of AI’.
  • In simple terms, AI refers to the ability of machines to perform cognitive tasks like thinking, perceiving, learning, problem solving and decision making. AI makes it possible for machines to learn from experience, adjust to new inputs and perform human-like tasks. 
  • AI functions on three basic concepts: machine learning, deep learning, and neural networks.

Read more here:


  • The datasets used to train AI are increasingly large and take an enormous amount of energy to run. This have an adverse environmental impact and such wasteful AI systems are called red AI.
  • The MIT Technology Review reported that training just one AI model can emit more than 626,00 pounds of carbon dioxide equivalent (CO2e).
  • MIT also reported that the cloud now has a larger carbon footprint than the entire airline industry, and a single data center might consume an amount of electricity equivalent to 50,000 homes.
  • According to Greenly’s estimates, the GPT-3 version of the AI service could have a footprint of some 240 tonnes of CO2e, the equivalent of 136 return flights between Paris and New York.
  • As a result of the pandemic, data and AI deployment increased exponentially as the demand for digital transformation increased.
  • Also, companies working with AI are hesitant to share data about their energy mix. Greenpeace’s Clicking Clean report from 2017 says that many companies who had committed to a 100% renewable future are more in a state of status quo than on a transformational path.


AI seems destined to play a dual role. On the one hand, it can help reduce the effects of the climate crisis, such as in smart grid design, developing low-emission infrastructure, and modelling climate change predictions. On the other hand, AI is itself a significant emitter of carbon.

  • Encourage Green AI:
    • Green AI, or the process of making AI development more sustainable, is emerging as a possible solution to the problem of power-hungry algorithms.
  • Rationalise AI’s use:
    • Machine learning is not always necessary to solve a problem. Developers should conduct a cost-benefit analysis before deploying ML.
  • Estimate carbon footprints of AI models.
    • Carbon accounting must be improved by delivering faster, more accurate data on carbon footprints and sustainability impacts.
    • Tools like Salesforce’s Net Zero Cloud, SustainLife, and Microsoft Cloud for Sustainability can help companies visualize and understand their missteps so they can spot opportunities for improvement.
  • Examine how and where data is stored:
    • Some of the biggest machine learning jobs might be moved to more carbon-friendly regions of the world. For example, companies can move their data centres to areas that run on hydroelectricity or renewable energy. It can make emissions vary by a factor of 40, between a fully renewable grid and a fully coal grid.
  • Increase transparency and measurement:
    • As AI researchers/companies publish their results/products, they should include measurements of how much energy was emitted during its development.
  • Increase research into green AI:
    • To make AI less polluting it needs to become more of a mainstream conversation. For this, academic research to analyse carbon footprint needs to be encouraged.
  • Adopt technological solutions:
    • Energy efficient systems, reuse of models instead of training them from scratch and use of more efficient Graphic Processing Units could be incentivized in the industry.
  • Develop regulatory framework:
    • Governments should consider setting up a regulatory frameworks and legislations in a way that would legally address transparency and sustainability in AI development.
    • Industrial standards, like other pollution standards, should be developed to ensure the mitigation of environmental impacts by AI.
    • Green AI certifications could be introduced to facilitate the industry process for promoting green AI development.

Best practice: Google’s “4M”

  • Google has identified four best practices that can significantly reduce energy and carbon emissions for anyone using Google Cloud services. These include:
    • Model: Selecting efficient ML model architectures, such as sparse models, can advance ML quality while reducing computation by 3x–10x.
    • Machine. Using processors and systems optimized for ML training, versus general-purpose processors, can improve performance and energy efficiency by 2x–5x.
    • Mechanization. Computing in the Cloud rather than on premise reduces energy usage and therefore emissions by 1.4x–2x.
    • Map Optimization: lets customers pick the location with the cleanest energy, further reducing the gross carbon footprint by 5x–10x.
  • By following these practices, Google claims, energy can be reduced by 100x and emissions by 1000x.



Q. While accelerating the adoption of AI and machine learning technologies, their environmental impact needs to be considered. Discuss.