At Wisk’s offices, Schrinner told me that air taxis could one day connect remote Indigenous communities to urban hospitals, or sightseers to the Bay Islands. “It’s only eighty kilometres from Brisbane to the Gold Coast,” he said. “It should only take an hour by car, but on a busy weekend it can take two.” Out of curiosity, I asked how long public transit takes. He paused before admitting that he didn’t know. But in a Wisk it might take as little as fifteen minutes. He picked up a toy model of Cora and said, “At back-yard barbecues, people tell me, ‘This is far-fetched, we’ve heard about flying cars for decades.’ But we saw it this morning, and it’s not a pie in the sky. I tell them, ‘It will happen.’ ”

People in the industry tend to think that flight is useful and awesome, and not necessarily in that order. One of the reasons that the idea of flying cars has endured is that it seems to promise two different kinds of freedom: on the one hand, to get from point A to point B without a lot of hassle; on the other hand, to know the euphoria of exploring the third dimension. Most people at these companies got into the business because they were personally enraptured by flight. They are nonetheless well aware that airplanes and automobiles have vastly different requirements, and that the vision of a car that both drives and flies never made a ton of sense. An inventor and professor named Paul Moller spent four decades working on his Skycar, believing for years that it was just within reach. In 2009, as the modern industry was starting, he went bankrupt. For a period, a company called ASKA had a mockup of a flying car in a window display in downtown Los Altos, California; members of its “Founders Club,” who paid a five-thousand-dollar deposit to get on the preorder list for an eight-hundred-thousand-dollar vehicle, were entitled to customize their aircraft’s design. Today, the storefront is vacant. (The company says it’s looking for a bigger showroom.) Soberer parties now understand that you have to make trade-offs between the breathtaking and the handy. During one of my visits to Wisk, I was inadvertently shown a set of talking points that employees had been given. Their comments were to be restricted to observations “grounded in reality/today,” and they were instructed to “avoid making bold claims that cannot be backed up with data/proof.” Beta was even reluctant to be portrayed as building anything like a flying car.

“A revolutionary war was brewing, but I managed to turn it into a culture war.”

Cartoon by Adam Douglas Thompson

One company already has a deal with United to start air-taxi services to Newark Airport next year, but such deals remain highly provisional: a German startup planned to fly passengers at this summer’s Paris Olympics, until European regulators quashed the idea. Dubai seems poised to begin offering services; China’s regulatory body just approved the mass production of an eVTOL for commercial use. But for such plans to become a widespread reality, let alone the future of transport, regulatory agencies will have to be coaxed into an overhaul of the way that airspace is structured and administered. We would need foolproof, digital detect-and-avoid systems to prevent collisions. eVTOLs might not be helicopter loud, but they are noisy. Engineers think that this issue can eventually be ameliorated, but, if it can’t, no one will be happy with an insistent mosquito buzz in the background. And then there’s the sheer number that would be needed. Before the pandemic, about four hundred thousand people a day crossed the Hudson River into Manhattan. Aerial commuting would require tens of thousands of drone taxis operating on regular, reliable schedules, with flawless safety records. David King told me, “So all of a sudden you’re into the realm of, ‘Why didn’t we just build a train?’ ”

Most responsible urban planners believe that public transit is a political problem, not a technological one. The tools to improve urban mobility—trains, trams, bicycles, sidewalks—have been around for a long time. The issue isn’t that we were promised flying cars and got a hundred and forty characters; it’s our attachment to a nostalgic kind of gizmo-first futurism, one that speaks to a profound failure of the national imagination. There’s something a little dismal about the fact that the mid-century dream of the future might, if everything goes perfectly according to plan, come to fruition in the form of saving half an hour en route to J.F.K. Then again, that isn’t too far from the initial vision. In an age of abundance, the promise of the future sold to affluent suburbanites was one of ever-greater consumer ease. George Jetson was not dizzyingly free; he had a flying car that folded up into a briefcase, and he used it to get to work quickly.

Almost everyone in the industry thinks that personal aerial vehicles will arrive eventually—maybe in twenty years, maybe in fifty. If they do, it will be through a series of gradual changes—the kind of thing that can look, to the untrained eye, like stagnation. Peter Thiel and J. Storrs Hall seem to find it almost personally insulting that jets look just like they did in the sixties. But Brian Yutko, a former Boeing executive who is now Wisk’s C.E.O., told me, “That’s so off base. Do the airplanes largely look the same if you’re a four-year-old holding up two pictures? Yes, but they’re actually pretty different. They got seventy per cent more fuel-efficient, they have a much longer range, and safety incidents have been driven as close to zero as possible.” He continued, “That’s technological progress that took human ingenuity for five or six decades. Other sectors should learn from that! It’s not stagnation, it’s ‘How did you do that?’ ”

Last fall, I joined Wisk’s team in New Zealand, where it was participating in an “airspace integration test”—a pioneering attempt to introduce unmanned flights into an area with commercial planes and other air traffic. New Zealand has a lot of empty sky, crashes pose a risk largely to sheep, and the country’s regulatory agencies have been hospitable to experimentation. The tests took place about twenty miles south of Christchurch, on a thin, dusty isthmus called the Kaitorete Spit. The airfield there is a joint venture with local Maori councils, who named it Tāwhaki, for a god who gathered knowledge from the heavens and brought it back to earth.

One design for early automobiles, known as Horsey Horseless, featured an artificial horse head mounted on the front of the chassis, so that it would resemble something familiar to other horses on the road. Wisk took a similar approach in the test. It was using a Boeing drone that was not being flown autonomously but by pilots in regular communication with air-traffic control; the pilots simply happened to be on the ground rather than in the cockpit. This was quite a big step for commercial aviation, in which virtually all procedures are based around the presence of an onboard pilot, but the accomplishment was buried under a barrage of bureaucratic acronyms: they hoped to do B.V.L.O.S. for an R.P.A. under I.F.R. conditions in C.T.A. One employee, who had come to Wisk from a jet-pack startup, told me that “a scaled-up U.A.M. operation would saturate the existing system within minutes.” U.A.M. means “urban air mobility,” and what they were doing here, he continued, was laying the groundwork for a P.S.U.

“What’s a P.S.U.?” I asked.

“A provider of services for U.A.M.,” he said, laughing. “It’s an acronym within an acronym.”

The pilots, dressed in vintage patterned fleeces and heavy work boots, were camped in the rear of a ground-control trailer. They were confident that their vehicle’s path would be even more precise and reliable than that of a standard craft. It was launched with the faint snow cover of the Southern Alps in the distance, and it climbed in a tight spiral to twenty-five hundred feet before the pilots requested clearance to enter controlled airspace. The reply came over the radio: “You are identified and cleared.” To a civilian, it wasn’t easy to tell what the fuss was all about. Aviators, however, knew better. Boeing’s regional executive, who had come to make a bottle-breaking appearance, declared it an “enormous milestone for aviation.”

Byron Airport is an uncontrolled airfield about forty miles east of the Bay Area. On the morning that I arrived, the winds were calm, at about three knots from the east, with scattered cloud cover at thirteen thousand feet. Mt. Diablo was visible in the haze, and the windmilled hillsides, freshened by a week of rain, were a deep green. Wyatt Warner, Pivotal’s chief test engineer, hauled a BlackFly on a trailer to a small landing pad, which made a sucking sound in the mud. We waited for a planeload of skydivers, who bloomed against the gray cloud cover, to finish their descent. Then Warner took off in the BlackFly for a nine-minute test flight. His flight was elegant, though the craft still looked as if it didn’t want to be in the air—like a tractor having a nightmare. Allison King, a mechanical engineer who was monitoring flight data, told me that she’d come across Pivotal’s Web site shortly after graduating from M.I.T. “I thought, Well, that’s just C.G.I.,” she said. “Then I looked at the disclaimer that said, ‘This is not C.G.I.,’ and I was, like, ‘Wait, what?’ ”

When Warner returned, Charlie Bushby, my first flight instructor, said, “You know what that means?” They gave me a flight suit with a smart BlackFly patch. Warner, watching the jumpers, was reminded of an old joke: “The good thing about when your parachute doesn’t open is that you have the rest of your life to solve the problem.” Idle conversation turned to plane-crash survival: a Serbian flight attendant who fell from thirty-three thousand feet and lived; those who walked away from the 1999 crash in Sioux City. We spoke about a local test pilot who had worked for Beta and interviewed at Pivotal; he had died weeks before in a tragic kit-plane crash. Bushby said, “Perhaps let’s not talk about this right now?”

The BlackFly was used in what may have been the first manned flight of an eVTOL.

With a clipboard checklist strapped to my thigh, I climbed up into the vehicle and slid the plexiglass canopy over my head. I turned the propellers on, tested the controls, and then initiated the takeoff sequence. My first flight was a simple hover—up and then down. The propellers began to spin with the sound of leaf blowers, the craft reared back, and I was wrenched upward, pinned to my seat, heart hammering. Warner had instructed me to breathe, but my body sent only lurching reminders that flight is wrong. I hung there just long enough to steal a hurried glance in each direction, then thumbed the toggle to descend. In what felt like both an instant and an eternity, I was back on the ground. We let the craft rest for ten minutes before my second test flight, which involved flying in a small box pattern over the landing pad. The flight plan meant a long interval in pure hover, and while I was airborne my motor temperatures quickly hit a hundred and twenty degrees. The yellow warning lights flashed; although I wasn’t perfectly centered over the pad, this seemed like a good time to land, which I did, with a little skid into the mud. I hopped out, and Warner and I dragged the BlackFly back to the proper takeoff spot.

I was now ready, they told me, to fly for real. As I took off into a hover, I twisted the stick to the right, turning away from everyone below, and lit out in the direction of the hills. Once I levelled off, the propellers quieted to a much softer hum, and all at once I had a feeling of lightness and agility in the air. Below me were muddy ponds, glistening patterns of water and grass, a cluster of black cows. I crested through a long, slow turn over the base of the foothills, and the machine felt alive to my touch. The company had disabled cruise mode, limiting me to an airspeed of about thirty miles an hour—something they said they’d done to prevent me from accidentally slipping into higher gear. But I knew now that, if cruise weren’t disabled, I would have pushed the craft to whatever speed was available to me; I would have flown in the direction of the hills and the sky and never come back; I would be up there still. There was the sense that the vibrating craft was an extension of my limbs. It was no wonder that the disembodied march of software had left so many technologists longing for the experience of living inside the circuitry. Entirely forgotten was the banality of commuting to work. The delirium of flight was enough. I glanced at my viewscreen, began the slow, gradual descent I’d practiced on the simulators, and, with great reluctance, triggered the sequence to land. ♦