The UK Is About to Move Planes 600 Metres to Stop Them Drawing Clouds
Operation Blue Skies is a 30-month trial using Google's AI forecasts to spot the patches of sky where contrails form, then nudging aircraft above or below them. Contrails cause about a third of aviation's climate impact.
Google, the UK air traffic service NATS, the Met Office, Contrails.org, Imperial College London and the University of Cambridge launched a programme on Tuesday called Operation Blue Skies. Over 30 months and backed by the UK Department for Transport, it will try something that has never been done across a whole ocean’s worth of airspace: predicting where aircraft will leave warming contrails, and routing them around those patches.
The setting is Shanwick Oceanic airspace, the eastern half of the North Atlantic, which handles a large share of Europe to North America traffic. Researchers estimate this one region accounts for around 5 percent of global contrail warming, most of it in winter. Two operational trials are planned, in the winters of 2026-27 and 2027-28, with roughly 20 to 40 test days each. On those days, controllers will ask selected flights to change altitude by up to 2,000 feet, about 610 metres, based on Google’s forecast of where the air is cold and humid enough for contrails to persist. Google is contributing 1.4 million pounds in kind, mostly the AI prediction models and satellite-based verification, within a programme costed at around 5 million pounds.
The physics is easier than it sounds. A contrail is a line of ice crystals that forms when hot, wet engine exhaust meets very cold air. Most vanish in seconds. A small minority form in thin bands of air that are already saturated with moisture, and those spread into cirrus sheets that hang around for hours, trapping heat that would otherwise escape at night. Those bands are thin, so a plane flying a few hundred metres higher or lower often misses them entirely. The awkward part is that they are hard to forecast, which is exactly the sort of noisy, high-dimensional pattern problem where machine learning tends to earn its keep. Estimates put contrails at roughly a third of aviation’s total climate effect, which is why a cheap altitude tweak is such an attractive lever.
Worth keeping expectations grounded. Flying higher or lower than optimum burns extra fuel, so the trial has to show the warming avoided beats the carbon added, and that arithmetic has been contested in earlier, smaller studies. Verifying the result is genuinely hard, because you are measuring clouds that did not form. And the whole thing is a research programme, not a policy: nobody is rerouting your holiday flight yet.
What this means for you: nothing you will notice on your next trip, and no reason to change how you book. What makes this one worth watching is the shape of it. Most AI climate stories are about a model that might one day help. This is a government, an air traffic controller and a weather service agreeing to change real flight paths on real days based on a model’s output, then publishing whether it worked. If the numbers hold up, it is one of the cheapest climate interventions in aviation. If they do not, we will find out in public, which is also useful.
Sources
Source: https://blog.google/innovation-and-ai/models-and-research/google-research/blue-skies/
ChatGPT Now Has a Separate Version for Teenagers, and It Guesses Who Needs It
OpenAI began rolling out ChatGPT for Teens on Tuesday. Anyone aged 13 to 17 lands in it by default, an age-prediction system routes uncertain cases there too, and parents can set quiet hours.
