May 14th 2025
Listen to this story
“THERE’S NEVER been anything even close to it—from speed to manoeuverability…to payload,” gushed Donald Trump, as he announced on March 21st that America’s future fighter jet, the F-47, would be built by Boeing, an aerospace giant. The jet is one of several so-called sixth-generation aircraft on drawing boards around the world.
In December China showed off what was believed to be a prototype of the J-36, an imposing plane with stealthy features and a large flying-wing design. Britain, Italy and Japan are co-developing their own plane, in Britain provisionally called the Tempest, which is due to enter service in 2035. France, Germany and Spain hope that their Future Combat Air System (FCAS) will be ready by 2040. Together, these represent the future of aerial warfare.
Fighter jets tend to be categorised by their age, features and sophistication. The first generation appeared in the 1940s and 1950s. Many of those in NATO service today, like America’s ubiquitous F-16, are fourth-generation ones, built from the 1970s to the 1990s. The latest fifth-generation planes, such as the F-35 and F-22, the latter perhaps the leading fighter jet in operation today, tend to enjoy stealth, the capacity for sustained supersonic flight and advanced computer systems.
By comparison with earlier planes, the sixth generation of jets all have one thing in common—they’re big. Early images of the F-47 have been heavily obscured and edited, and might bear little resemblance to the final plane. But photos of the J-36 and models of the Tempest (pictured) indicate aircraft far larger than the fourth-generation Chinese J-20 and European Typhoon or fifth-gen American F-35 and F-22. The similarity suggests that all these countries have similar prognoses about the future of war in the air.
One shift they all predict is more, and better, surface-to-air missile systems, a lesson reinforced by the strong performance of air defences in Ukraine. That requires more stealth to keep planes hidden from enemy radar. Stealth, in turn, requires smooth surfaces—bombs and missiles cannot hang off the wing, but must be tucked away inside a larger body.
A second shift is in the increasing range of air combat. For the past 40 years, the proportion of air-to-air kills that occur “beyond visual range” has grown steadily—from a tiny fraction of all in the 1970s to more than half between 1990 and 2002. Since then air-to-air missiles have been able to travel ever farther. Europe’s Meteor, with a 200km range, was at the forefront of technology when it was first tested a decade ago. America’s AIM-174B and China’s PL-17 can now hit things 400km away. That means planes need better sensors to spot and fire at targets from farther away; they also need better electronic warfare equipment to parry incoming threats. These technologies require more space to generate power and remove all the heat that electronics tend to produce.
Finally, planes are especially vulnerable to long-range missiles when they are on the ground. That means they need to fly from more distant airfields, requiring larger fuel tanks and less drag for more efficient flight. The huge wings seen on the Tempest and the J-36 allow for both those things, notes Bill Sweetman, an aviation expert. Range is a particular concern for America. Its airbases in Japan are within reach of vast numbers of Chinese ballistic missiles. It plans to disperse its planes more widely in wartime and to fly them from more distant runways, such as those in Australia and on Pacific islands.
Top-flight
Selected fighter jets in service, 2025
Selected sixth-generation, due in service 2030s
F-47
Wingspan 14.25m
Length 19.2m
F-35
Wingspan 10.7 metres
Length 15.7 metres
F-35
Typhoon
F-22
F-47
Tempest
J-36
Design unknown
Combat radius, km
1,093
1,852*
2,222*
Unknown
1,240
1,389
Sources: Lockheed Martin; BAE Systems; Boeing; media reports
*Estimate
Long-range planes are appealing for several reasons. “We’re talking about really extreme ranges,” notes Group Captain Bill, the Royal Air Force (RAF) officer in charge of thinking through how the service will use the Tempest, speaking recently (without his surname) on the “Team Tempest” podcast, which is produced by the consortium building the aircraft. The plane will need to be able to cross the Atlantic Ocean on a single tank of fuel, he says, a journey that would require today’s Typhoon jet to be refuelled three or four times. One reason for that might be that big refuelling tankers, which once sat safely to the rear of the front line, are increasingly vulnerable to new air-to-air missiles, like China’s PL-17. Another is that the Tempest could then take circuitous routes, avoiding Russian air defences along the obvious paths.
Put all this together and you get planes that look like old-fashioned bombers. Mr Sweetman compares the hulking J-36, with massive wings and cavernous weapon bays, to an “airborne cruiser”, optimised for range, stealth and carrying capacity over dogfighting agility. The single most important requirement for the Tempest is the ability to carry a lot of weapons, says Group Captain Bill, noting that it will have roughly double the payload of the beefiest F-35. That makes sense: if you can deliver more firepower per sortie, you can destroy a target with fewer risky flights into enemy airspace. “The same answers tend to pop up for all,” says Mike Pryce, who has advised Britain’s defence ministry on combat air design. “Stand off, don’t be seen, shoot first, don’t get into a knife fight.”
As the planes get bigger, their insides are also evolving into what are essentially “flying supercomputers”, says Roberto Cingolani, the CEO of Leonardo, an Italian company that is developing the wider Tempest programme along with Britain’s BAE Systems and Japan’s Mitsubishi. Leonardo says that the Tempest will be able to “suck up” a medium-sized city’s worth of data in one second, according to Tim Robinson of the Royal Aeronautical Society. That could include anything from radio traffic to the emissions of air-defence radars. The point is to share that data with friendly forces, including tanks and ships, says Mr Cingolani, perhaps via satellite, with a “central artificial intelligence” making decisions—presumably which targets should be attacked, by what, and when. Some might suggest “that’s science fiction,” he says. “No, that’s a vision.”
Perhaps the most contentious design choice is whether sixth-generation planes should have pilots. Elon Musk, Mr Trump’s aide, recently mocked the fact that “Some idiots are still building manned fighter jets.” In practice, most air forces believe that artificial intelligence (AI) and autonomy are not yet mature enough to allow a computer to replace a human pilot entirely; that will take until 2040, reckons the RAF. Images of the F-47, though unreliable guides to the final product, depict “a relatively large bubble canopy”, notes Thomas Newdick of the War Zone, a website, “providing the pilot with excellent vision”. Some missions are particularly sensitive: France will use the FCAS to deliver nuclear weapons, a task that may always remain a human prerogative.
Nevertheless, the prevailing idea is that sixth-generation planes will be the core of a larger “combat air system”, in which a human in the cockpit controls a larger fleet of uncrewed drones, known, in American parlance, as collaborative combat aircraft (CCA). “The concept is that you have an aircraft-carrier that is flying,” says Mr Cingolani. “It’s an entire fleet that moves in the sky and makes decisions.” The human in the cockpit is best described not as a pilot, says Group Captain Bill, but as a “weapons system officer”, the RAF’s term for someone managing sensors and weaponry.
On May 1st America’s air force announced that it had begun ground testing its two CCA prototypes in advance of flight tests later this year. Current order numbers suggest that each F-47 will get two CCAs. The drones might scout ahead, spot targets or carry weapons themselves—all within line-of-sight and under “tight control”, notes Frank Kendall, a former air-force secretary. Much of the intensive computing required to carry out these tasks will need to take place on board the crewed mothership, with relevant data shared to all craft instantaneously, says Mr Cingolani, speaking in the context of the Tempest. He emphasises that the communication links have to be secure. “I’m not sure in ten years we can make it.”
If he and his company can pull it off, it will cost a pretty penny. Mr Kendall, in the Biden administration, paused the development of the F-47 in large part because it was expected to cost twice as much as the F-35—perhaps as much as $160m-180m apiece—which would mean the government could afford only a small fleet of 200 or so planes. Many in the Pentagon wanted a greater emphasis on building CCAs to complement the existing fleet of F-35s, rather than pouring money into a new platform that might not turn up until long after a war with China.
In Britain, Justin Bronk, an air power expert at the Royal United Services Institute, expresses similar concerns, drawing an analogy with the experimental versus war-winning weapons of the second world war. “Pouring all the money that defence can spare…into a programme that, in the best case, will not deliver a fully operational capability before 2040 feels to me like the UK concentrating all Air Ministry resources on Avro Vulcan development in 1936,” he says, citing a plane that did not appear until a decade after the war was over, “rather than Hurricanes, Spitfires, Blenheims, Whitleys and Wellingtons.” ■
Curious about the world? To enjoy our mind-expanding science coverage, sign up to Simply Science, our weekly subscriber-only newsletter.
Science & technologyTechnology
This article appeared in the Science & technology section of the print edition under the headline “Flying supercomputers”
→Britain is now the biggest funder of solar-geoengineering research
→For the first time, a CRISPR drug treats a child’s unique mutation
Discover stories from this section and more in the list of contents
The Economist today
A daily newsletter with the best of our journalism
Yes, I agree to receive exclusive content, offers and updates to products and services from The Economist Group. I can change these preferences at any time.