Aviation’s climate challenge may have found an unexpected ally – artificial intelligence.
American Airlines and Google have revealed promising results from a joint trial using AI-powered forecasting to significantly reduce contrails – a major but often overlooked contributor to global warming.
Contrails, or condensation trails, form when aircraft fly through cold and humid air, creating ice crystal clouds around engine emissions. These clouds can trap heat in the atmosphere, amplifying the warming effect of aviation.
To tackle this, Google developed an AI system capable of predicting where contrails are most likely to form. American Airlines integrated these forecasts into its flight planning systems, allowing pilots to make small adjustments – such as changing altitude or route – to avoid high-risk zones.
The approach is gaining attention as one of the most cost-effective and scalable climate solutions currently available to the aviation industry, which faces growing pressure to reduce its environmental footprint.
According to Contrails.org, contrails are responsible for roughly 1% to 2% of global warming. The organization, part of the Breakthrough Energy initiative founded by Bill Gates, collaborated on the trial.
Unlike sustainable aviation fuel – which remains expensive and limited in supply – avoiding contrails can often be achieved through minor operational changes. While some contrails dissipate quickly, others can persist for hours or even days under highly humid conditions, making targeted avoidance particularly impactful.
The trial covered 2,400 flights between the United States and Europe. Half were given AI-informed routing options, while the rest operated as usual. Among 112 flights that followed the adjusted routes, contrail formation dropped by 62% compared to the control group.
Researchers estimate this translated into a 69% reduction in the flights’ warming impact.
Dinesh Sanekommu, who leads Google’s contrail research, said the findings demonstrate how AI can play a practical role in decarbonizing one of the hardest sectors to transform.
As airlines search for realistic pathways to cut emissions, the results suggest that smarter routing – powered by data and machine learning – could become a key tool in reducing aviation’s climate footprint without major cost increases or operational disruption.
