Civil Aviation: Google Unveils a New AI to Tackle Contrails

August 31, 2026

Over the next two years, Google will test a new artificial intelligence in collaboration with the UK government. This AI will be tasked with proposing small trajectory changes to airlines’ aircraft to limit contrails whose impact on the climate is well established.

Two Major Tests Ahead

First, it should be noted that the contrails (cirrus clouds) from civil aviation have a significant warming effect, often regarded as comparable to or even greater than the sector’s CO₂ emissions. The problem lies in the fact that these persistent trails act as a blanket trapping the infrared radiation emitted by the Earth and thus warm the atmosphere. Unfortunately, non-CO₂ effects of civil aviation—largely stemming from contrails—account for nearly two-thirds of the sector’s total climate footprint.

In this context, the tech giant Google and the UK government have officially launched a broad experimental program, as revealed in a blog post published on August 18, 2026. This initiative, titled Operation Blue Skies, will aim to test over two and a half years the avoidance of contrails across an entire airspace. The tests will cover the North Atlantic corridor in the Shanwick area, a route through which thousands of flights between North America and Europe pass each year.

Before any potential democratization of the system, Google will conduct two large-scale tests during the winters of 2026-2027 and 2027-2028. Researchers will monitor the traffic of about 10,000 aircraft, but only some of these flights will undergo trajectory changes. In practice, pilots will need to make minor adjustments, that is, small altitude changes—typically up to 2,000 feet (600 m)—following air traffic control instructions.

Artificial Intelligence and Quantified Goals

The heart of the project is a new artificial intelligence developed by Google. It will combine massive volumes of weather data and predictive models to pinpoint exactly where cold, humidity-saturated air conducive to persistent contrails lies. After proposing trajectory changes, the AI will use machine learning and satellite imagery to monitor the sky in real time and confirm whether the changes have or have not prevented contrail formation.

According to Google, the deviations will cause a slight increase in fuel consumption, about 100 kg per modified flight. By contrast, avoiding contrail formation should yield an impact equivalent to several tons—or tens of tons—of CO₂ saved per flight. If the tests prove successful, Google would eventually develop a standardized operational guide that could be applied to all airspaces on the planet.

A broad collaboration with a range of British bodies

The collaboration with the United Kingdom is valuable, as several renowned actors are taking part in the project. Notably, the National Air Traffic Services (NATS), the country’s main air traffic services provider, will be responsible for safety and for communicating the deviation instructions to pilots. The British meteorological service Met Office will lend its expertise to ensure the reliability of atmospheric forecasts.

The avoidance of contrails at the scale of airspace requires close collaboration across scientific, technological, and air traffic management domains. Operation Blue Skies brings together a world-renowned British consortium with the expertise needed to turn cutting-edge research into operational reality., reads the blog post.

Finally, we should mention the scientific organization Contrails.org for shaping the parameters of the tests or, moreover, several experts from the University of Cambridge and Imperial College London for analyzing the results and the quantification of the real climate benefit.

Sindre Halvorsen

I write about space exploration, frontier science and the technologies that are quietly shaping the future. From Norway, I follow the missions, discoveries and ideas that connect life on Earth with what lies beyond it. My goal is to make complex subjects clear, useful and worth paying attention to.