Can AI Tackle Aviation's Climate Problem?

Those wispy white lines behind a plane aren't harmless. Google and American Airlines just tested whether AI can steer flights around them.

Welcome Back,

Look up at a clear sky and you'll probably spot a few thin white streaks trailing behind distant planes. They look almost decorative, but those contrails are quietly doing more climate damage than the carbon dioxide those same planes burn to create them.

On March 19, 2026, American Airlines and Google announced results from a real-world trial using an AI-driven forecasting tool to help pilots dodge the exact atmospheric conditions that create contrails, calling it one of the most scalable climate solutions currently available in aviation, according to the Associated Press.

Today we look at how contrails actually warm the planet, what the AI in this trial specifically predicts, the real numbers behind the 2,400-flight experiment, and why aviation keeps getting singled out as one of the hardest industries to decarbonize.

☁️ Why Contrails Are a Bigger Deal Than They Look

Contrails form when aircraft fly through cold, humid air, letting ice crystals build around the soot particles engines expel, which then form clouds that trap heat and contribute to global warming, according to the Associated Press.

The scale of that effect is surprising. Contrails contribute a striking 1% to 2% of all global warming, according to Contrails.org, a nonprofit dedicated to aviation climate mitigation that's part of Bill Gates' Breakthrough Energy initiative, per the same AP report.

Put another way, the 2022 IPCC report found that clouds created by contrails account for roughly 35% of aviation's total global warming impact, which is actually more than half the warming effect of the jet fuel those planes burn, according to the World Economic Forum.

Not every contrail matters equally. Many are short-lived and dissipate quickly, but under extremely humid conditions, some can persist and spread for hours or even days, turning a thin line into a lingering, heat-trapping cloud cover, per the AP.

Contrails form when soot from jet engines meets cold, moist air at cruising altitude, creating heat-trapping ice clouds.

🛰️ How the AI Forecasting Tool Actually Works

Google's approach uses AI to predict where contrails are likely to form as planes fly through specific atmospheric conditions, then feeds those forecasts into American Airlines' flight planning system, letting pilots adjust altitude or route to avoid the danger zones, according to the Associated Press.

The project, called Operation Blue Skies, brought together Google, American Airlines, and other collaborators specifically to test whether small, targeted route changes could meaningfully cut contrail formation without major disruption to flight operations, according to BBC's Tech Life.

That's a meaningfully different strategy from the industry's other major climate lever, sustainable aviation fuel, which can cut emissions and contrails significantly but is far more expensive to deploy than simply adjusting where a plane flies, according to the AP.

📊 What the 2,400-Flight Trial Actually Found

The trial covered 2,400 flights traveling from the U.S. to Europe. Half were given an alternative route designed to avoid contrail-forming conditions, while the other half flew their normal path as a control group, according to the Associated Press.

Of that group, 112 flights actually took the AI-recommended alternative route, and those flights saw a 62% reduction in contrail formation compared to the control group, with researchers estimating a 69% decrease in the associated warming effect, per the same report.

American Airlines and Google confirmed the same figures in a joint announcement, describing it as a statistically significant reduction in contrail formation and a substantial cut in related warming, according to American Airlines' newsroom.

The trial found that avoiding contrail-prone airspace made a measurable dent in warming without a meaningful fuel penalty.

⛽ The Fuel Cost Fear That Didn't Materialize

The biggest reason airlines have historically avoided rerouting for contrails is cost, altering a flight path typically burns more fuel, and fuel is one of an airline's largest operating expenses. This trial was designed specifically to test that assumption, according to the Associated Press.

The result surprised even the researchers: there was no statistically significant difference in fuel consumption between the flights that took the AI-recommended detour and the control group that flew normally, per the same AP report.

Dinesh Sanekommu, who oversees Google's contrail initiatives, framed the goal as building an evidence base rather than declaring victory outright, saying the hope is that AI-driven forecasts and operational trials like this one contribute valuable data to help the industry make informed decisions over time, according to the AP.

⚖️ Contrail Avoidance vs. Other Aviation Fixes

Contrail forecasting is one piece of a much bigger aviation decarbonization puzzle, one where different fixes operate on very different timelines and require very different investments.

On the fuel side, American Airlines and Google separately signed a record-breaking sustainable aviation fuel agreement in June 2026, unlocking 35 million gallons of SAF over three years and an estimated nearly 300,000 metric tons of CO2 equivalent emissions reductions, according to American Airlines.

Aviation contributes roughly 2 to 3% of global CO2 emissions, and the industry is under growing pressure to find scalable near-term fixes.

🧠 AI Spotlight Analysis

Aviation is one of the toughest sectors to decarbonize because there's no easy substitute for jet fuel at scale yet, and fleet replacement cycles run decades. That's exactly why an operational, AI-driven fix that works with today's planes and today's fuel is such an attractive first move.

It's worth keeping the scale in perspective too. Aviation contributes 2% to 3% of global CO2 emissions overall, according to American Airlines, meaning even a major win on contrails only chips away at a slice of a much bigger climate picture, one that still requires SAF, fleet upgrades, and policy alongside AI-driven operational fixes.

The bigger test now is scale. A 2,400-flight trial on one transatlantic corridor is promising, but rolling contrail forecasting out across global airspace, with different weather patterns, air traffic control systems, and airline fleets, is a very different engineering and policy challenge.

🔗 Sources and Further Reading

Thanks for reading AI Spotlight.

Our mission is simple: deliver clear, trustworthy, and actionable AI insights that help professionals stay ahead without the hype.