Play store
Artificial intelligence (AI) is emerging as a potential tool in the fight against climate change, offering applications ranging from improved weather forecasts to more efficient renewable energy systems, despite growing concerns over the carbon footprint of data centres.
The issue is expected to feature prominently at the COP31 climate summit in Turkey next month, where the hosts plan to promote “climate-responsible AI”.
Debates over AI’s environmental impact and concerns about its potential risks have often overshadowed its role in tackling climate change. However, experts say the technology can help reduce energy consumption, monitor industrial emissions and improve climate resilience.
Many climate-related AI applications consume less energy than general-purpose language models such as OpenAI’s ChatGPT and Anthropic’s Claude, according to experts.
“Data centres have come in as a challenge,” Siddharth Singh, an AI expert at the International Energy Agency (IEA), told AFP, noting concerns about the technology’s implications for energy systems and the climate.
“At the same time, because (AI) is a tool, it’s an opportunity. It really comes down to how it is deployed,” he said.
Improving renewable energy efficiency
AI is already helping operators manage complex electricity grids by improving forecasts of solar and wind power generation.
By predicting cloud cover, sunlight and wind conditions more accurately, the technology enables energy companies to anticipate renewable electricity output and reduce curtailment — when available power is cut because the grid cannot accommodate it.
According to the IEA, AI deployed across transport, industry and buildings could save 13.5 exajoules of energy by 2035, equivalent to Indonesia’s annual energy demand.
AI could also reduce emissions by scheduling electricity-intensive activities, such as running washing machines, when solar power is abundant, Singh said.
Environmental technology nonprofit WattTime helps companies schedule electric vehicles, phones and other devices to operate when electricity comes from cleaner sources.
“One of the things AI is very good at is forecasting,” said WattTime co-founder Gavin McCormick.
Transforming weather forecasting
AI is also reshaping weather forecasting by producing high-quality predictions much faster and at lower costs.
Florian Pappenberger, director-general of the European Centre for Medium-Range Weather Forecasts (ECMWF), said AI had significantly improved the speed and performance of weather prediction.
He said AI-based forecasting remained grounded in physics-based models while enabling the centre to achieve in two years progress that would traditionally have taken a decade.
Lower costs also allow the centre to generate more forecasts, while the technology can strengthen early-warning systems for extreme weather events.
Google says one of its AI systems can forecast river floods up to seven days in advance across more than 150 countries.
Rather than replacing human expertise, Pappenberger said, AI could enhance people’s capabilities and help them perform tasks more effectively.
Tracking greenhouse gas emissions
AI and satellite technology are also being used to identify greenhouse gas emissions from individual facilities, including oil and gas installations, mines and landfills.
Climate TRACE, an initiative co-founded by McCormick, uses AI to monitor emissions worldwide and compile inventories of pollutants.
“Climate TRACE is 20 years more advanced because of AI,” McCormick said.
He added that the initiative had made it possible to provide emissions inventories to cities, helping many of them develop climate action plans.
Fossil fuel concerns remain
Despite its potential to support clean energy, AI has also been used extensively by oil and gas companies to analyse geological data, identify underground reservoirs and improve fossil fuel production.
Maria Joao Sousa, executive director of Climate Change AI, a nonprofit promoting the use of AI for climate action, warned that such applications could have a negative impact on the climate.
AI’s environmental consequences will therefore depend partly on how the technology is deployed, experts say.
“AI will never be a silver bullet,” Sousa said. “Technology itself will never be the solution alone.”