Google is partnering with the UK Government and aviation industry on Operation Blue Skies, a major trial using AI to help aircraft avoid atmospheric conditions that produce strongly warming condensation trails, or contrails. Contrails form when water vapor in jet exhaust condenses around particles at high altitude. Although they look like thin clouds, some persist and trap heat, and Google says contrail warming accounts for roughly one-third of aviation's total climate impact.
The project builds on Google's earlier research showing that AI can forecast where warming contrails are likely to form. Instead of requiring major changes to aircraft or engines, the system can identify areas of airspace where a small change in flight altitude or route could prevent a persistent contrail. Google says previous trials with airlines and air-navigation organizations demonstrated that flight crews and controllers could make these targeted adjustments while remaining within normal operational procedures.
The new programme moves from individual flight experiments to airspace-scale coordination. Operation Blue Skies will operate in Shanwick airspace, covering the eastern half of the North Atlantic corridor. Google estimates that this region represents about 5% of global contrail warming. Over the 30-month programme, two operational trials of approximately four months each will be conducted, with around 10,000 flights passing through the trial area during trial hours each year. Only a small percentage of flights expected to create significant warming contrails would be slightly diverted.
The project is notable because it combines AI prediction with real-world infrastructure and independent measurement. Google is providing £1.4 million in-kind through AI expertise, engineering and computing resources, while NATS will handle air-traffic operations and safety assessment. Imperial College London and the University of Cambridge will independently evaluate the climate impact, and the UK's Met Office will contribute meteorological forecasting. This makes the initiative more than an AI demonstration: it is an attempt to establish whether machine-learning predictions can be translated into safe, measurable climate action across an entire aviation corridor.
The larger significance is that AI may help address climate problems without waiting for entirely new physical infrastructure or aircraft technology. If relatively small operational decisions can prevent a disproportionately large climate effect, AI could become a practical optimization layer for existing transportation systems. Operation Blue Skies will ultimately test whether that promise survives the complexities of real-world air traffic, safety, weather uncertainty and independent climate measurement.
