Positive AI for the environment
Arguably, AI can also be utilised to have a positive impact on the environment.
AI can process huge volumes of data and is increasingly efficient, producing interpretable results from large datasets extremely quickly. There are thus ways in which AI can be used to benefit the environment.
In the 2024 blog post, “The Truth About AI & The Environment: Pros and Cons“, published by Astutis, Enock Ebbah explains:
“Leveraging AI can transform traditional grids into intelligent, adaptive, efficient, and sustainable energy systems by implementing AI algorithms based on machine learning, data analytics and the collective network of connected devices across the internet”.
For example, the ExaGeo PhD project is exploring the use of AI and machine learning methods to enable some of these infrastructure changes across the UK, harnessing Exascale computing.
TransiT is another UK-based initiative using AI, machine learning, and digital twinning to help decarbonise the UK transport sector.
TASK: Take a more in-depth look at one of the projects from TransiT or ExaGeo. How would you expect the work to impact your day-to-day life in the near to mid future?
AI might also be beneficial to some environmental sustainability projects. This would involve managing and reducing environmental impacts across the lifecycle of a system to ensure present needs are met without compromising the ability of future generations to meet their own needs.
TASK: Listen to The Good Robot podcast, episode “Surfing the Web in Sync with the Sun with Anne Pasek” co-hosted by Dr Kerry McInerney and Dr Eleanor Drage.
In this episode, Dr Anne Pasek (Canada Research Chair in Media Culture and the Environment) discusses why addressing the climate crisis doesn’t necessarily mean living with less, but rather learning to live in sync with nature.
Do you think the practical approaches to AI and sustainability discussed in the podcast could make a real difference? Can you think of any behaviours you might change to support AI for sustainability?
The 2025 report “Engineering Responsible AI: foundations for environmentally sustainable AI” (previously discussed) proposes five steps to reach environmentally sustainable AI:
- Expanding environmental reporting mandates
- Addressing information asymmetries across the value chain
- Setting environmental sustainability requirements for data centres
- Reconsider data collection, transmission, storage and management practices (i.e. the entire lifecycle of AI systems)
- Leading the way with government investments
In summary, here are just some of the ways in which AI can improve the environment and sustainability:
- Optimising carbon capture – AI can enhance the effectiveness of carbon capture through improving computation and mitigating weaknesses of data-driven models.
- Environmental monitoring – AI can process large volumes of satellite imagery to assess land-use changes, deforestation and pollution levels. These models can also be used to predict natural disasters.
- Implementation with renewable energy – AI models allow the effective integration of renewable energy sources like solar and wind into the national electricity grid.
- Efficiency of systems – AI can improve the efficiency of systems in areas like carbon capture, plastic recovery, energy, and recycling schemes, thus reducing waste and energy requirements.
Further reading links for environmentally sustainable AI:
https://www.sciencedirect.com/science/article/pii/S0048969723025342
https://blog.cleanhub.com/how-is-ai-being-used-to-help-the-environment
