Artificial intelligenceEnvironment

Google uses AI to avoid aircraft contrails and reduce their warming effect by about 40%. From data center water consumption to particles in the sky.

8 September 2026

The white trails planes leave in the sky are not a cosmetic detail: they contribute concretely to global warming. And according to the research Google cites, they account for roughly a third of aviation's entire climate impact.

Google is testing a system based on artificial intelligence that helps aircraft avoid the parts of the atmosphere where persistent condensation trails are most likely to form. The results obtained with Cathay Pacific are clear: according to Google's satellite analysis, the flights that followed routes designed to avoid contrails achieved a reduction of about 40% in the warming effect caused by condensation trails.

Why aircraft contrails warm the planet

Condensation trails, also called contrails, form when the water vapour in engine exhaust meets cold, humid air at high altitude.

Many vanish quickly. Some, instead, persist for hours, spread out and turn into genuine artificial clouds.

These clouds trap heat in the atmosphere and push temperatures up. It is not a marginal effect: the research Google cites estimates that persistent contrails weigh in for around a third of the total climate impact of air travel. Put plainly, a huge share of the climate damage from planes does not come from the CO2 leaving the engines, but from the clouds those engines create.

This is exactly what Project Contrails, Google's project dedicated to reducing contrails with artificial intelligence, is aimed at.

How Google's AI avoids contrails

The principle is simple: stay clear of the bands of atmosphere where temperature and humidity favour the formation of persistent trails.

To do it, Google's system combines artificial intelligence models, satellite imagery, advanced weather forecasts and data on routes and flight altitudes. The AI identifies the risky areas in advance and lets pilots and dispatchers plan small changes of altitude or route.

There is no need to overturn the aircraft's path. In many cases a limited change of altitude is enough to dodge an air mass that is particularly prone to contrails. Sometimes, to significantly cut the warming produced by a flight, it can be enough to change altitude by a few hundred metres.

The trial with Cathay Pacific

Cathay Pacific is the first airline in Asia-Pacific to work with Google on this project, and the first in the world to test contrail avoidance on ultra-long-haul flights.

The airline ran an operational trial on more than 100 flights across its network, including connections of over 16 hours and more than 10,300 kilometres. More than 80 of those flights managed to follow routes designed specifically to avoid the areas most favourable to contrail formation.

And here is the number that matters. According to Google's analysis of the satellite imagery, those flights achieved an estimated reduction of about 40% in the warming effect produced by condensation trails.

It has to be stated precisely, because the number is strong and needs to be understood properly: it does not mean the planes' overall emissions dropped by 40%, nor that aviation cut its total contribution to warming by 40%. The figure concerns the climate effect of the contrails on the flights involved in the trial. Which, given how much contrails weigh, is still an enormous intervention.

The key point: it works with today's planes

Here is the most interesting part, and it should be said without hedging. This strategy does not require new aircraft, revolutionary engines or next-generation fuels.

It applies to the planes already in service, simply by changing flight planning according to atmospheric forecasts. It is one of the very few measures to cut aviation's climate impact that could be adopted now, without waiting twenty years for fleets to be renewed.

Of course every change of altitude or route has to be weighed against fuel consumption, safety and traffic management: the climate benefit has to be large enough to justify any extra fuel. But when the maths works out, the benefit is immediate.

Google wants to widen the trials

After the results with Cathay Pacific, Google announced the extension of the trial to a larger number of flights, on polar, transpacific and intra-Asia routes.

The company is working with airlines, air traffic authorities, flight-planning software makers and with Contrails.org, the non-profit specialised in studying and mitigating contrails.

To this is added Operation Blue Skies, the British consortium announced in August 2026 that Google is part of, together with the Met Office, NATS, Contrails.org, Imperial College London and the University of Cambridge, backed by the UK Department for Transport. It is a 5 million pound, 30-month programme that will bring contrail avoidance to a wide portion of North Atlantic oceanic airspace, where a significant share of global contrail warming passes through.

What this news really tells us

It is one of the most concrete uses of artificial intelligence seen so far: analysing atmospheric and satellite data to make operational decisions that directly reduce a phenomenon that heats the planet.

Do not misread it, though. It does not solve the problem of planes' CO2 and it is not the decarbonisation of aviation. But it has a merit almost no other measure has: it can be applied now, with the planes and fuels we already have.

If a result like the 40% from the Cathay Pacific trial held up at scale, route planning in the future would no longer look only at safety, time and fuel, but also at the climate impact of the air the plane flies through. And sometimes, to weigh far less on the climate, it would be enough to climb or descend by a few hundred metres.

And here it gets interesting

There is, though, a question many people ask and that, at the level of public communication, nobody has yet answered convincingly: why, on the same route and altitude, or even intermittently along the same flight, do some planes leave long persistent trails and others none? A technical explanation exists (humidity and temperature change even within a few kilometres, and different engines and fuels produce different amounts of soot), but it is an explanation that does not satisfy many, and for years the matter was dismissed rather than explained.

The cultural path we have taken is curious too: we went from denying that certain atmospheric phenomena had any effect at all, to first having to justify them and then normalise them. Today aircraft contrails are officially a serious climate problem, to the point that Google builds an AI system on top of them.

And here is the point worth keeping firmly in view. While Google spends resources and AI to make planes avoid contrails, because they warm the planet, there are companies starting to experiment with the exact opposite: deliberately injecting particles into the upper atmosphere. It is so-called solar geoengineering.

One example is Stardust (stardustsrt.com), whose "Sunlight Reflection Technology" proposes dispersing particles in the stratosphere to reflect part of the sunlight and, in their words, "stabilize global temperatures in a safe, controllable and reversible manner". In practice, columns of trails that in their view would serve to cool the planet: the exact opposite of what Google attributes to aircraft contrails.

The tone is striking. The site is full of unrequested reassurances: "responsible development", a technology that is "safe, rapid and reversible" and "mimics a naturally occurring phenomenon". And when someone insists so hard on reassuring you about something you never asked about, it is usually the moment to raise your attention, not lower it.

Put together, the picture is bizarre to say the least: on one side artificial intelligence is used to take trails out of the sky because they warm, on the other there are plans to put new ones up because they would cool. Above our heads, with the atmosphere we breathe used as the laboratory, and with public communication that for years pretended the sky was not a topic. Both things are worth watching with our eyes wide open.