Bloomberg – How Apple Sank About $1 Billion a Year Into a Car It Never Built
Tim Cook shut down plans to acquire Tesla before cycling through a junkyard’s worth of self-driving designs over the past decade. The inside story is a case study in indecision.
Editors note: The Bloomberg article describes how Apple tried to outdo everyone else and start with Level 5 autonomy. That was one of their many mistakes. Also, interesting that Steve Jobs was apparently first to raise idea of an Apple car. Also, Apple considered buying Tesla at one point, and considered partnering with various automobile companies to actually manufacture the vehicle. But the biggest problem was indecision.
See also:
Original article by Mark Gurman and Drake Bennett from Bloomberg
Around the beginning of 2020, Apple Inc.’s top executives gathered at a former Chrysler testing track in Wittmann, Arizona, to try out the latest incarnation of the car the technology giant had been trying for years to make. The prototype, a white minivan with rounded sides, an all-glass roof, sliding doors and whitewall tires, was designed to comfortably seat four people and inspired by the classic flower-power Volkswagen microbus. The design was referred to within Apple, not always affectionately, as the Bread Loaf. The plan was for the vehicle to hit the market some five years later with a giant TV screen, a powerful audio system and windows that adjusted their own tint. The cabin would have club seating like a private plane, and passengers would be able to turn some of the seats into recliners and footrests.
Most important, the Bread Loaf would have what’s known in the industry as Level 5 autonomy, driving entirely on its own using a revolutionary onboard computer, a new operating system and cloud software developed in-house. There would be no steering wheel and no pedals, just a video-game-style controller or iPhone app for driving at low speed as a backup. Alternately, if the car found itself in a situation that it was unable to navigate, passengers would phone in to an Apple command center and ask to be driven remotely.
In the Arizona desert, Chief Executive Officer Tim Cook, Chief Operating Officer Jeff Williams and senior members of Apple’s design team sat in the prototype as it drove itself around a test track. They loved what they saw. But there was a catch, as the car project’s head, Doug Field, made clear: A lot of work still needed to be done before the autonomous driving system would work in the real world. Field, who’d been hired away from Tesla Inc. to oversee the project, proposed scaling back the self-driving goals to Level 3, which requires a human driver to be ready to take over at a moment’s notice, not watching TV or FaceTiming in a backward-facing seat. But Field’s bosses wanted Level 5.
The next year, Field left Apple to take over the electric-vehicle and software engineering efforts at Ford Motor Co. Under Field’s successor, Kevin Lynch, who also runs Apple’s smartwatch software group, the car’s design continued to evolve. It had become pod-shaped, with curved glass sides that doubled as gull-wing doors, and the company considered including ramps that would automatically fold out to make heavy cargo easier to load. The front and the back were identical, and the only windows were on the sides, a design choice with potentially dire consequences in the event that a human needed to do any driving. (Front and rear windows were later added.) Some people on the project called it the I-Beam.
The I-Beam never made it into production, nor did any of Apple’s other designs. Now, it seems, they never will. On Feb. 27, Apple told staff it was giving up on developing a car. That decision, while abrupt, was not a surprise. Over the past decade, the company toiled away on at least five different major designs, drove prototype self-driving systems for more than a million miles, hired engineers and designers only to lay them off, and weighed partnerships or acquisitions with Tesla, Mercedes-Benz, BMW, Volkswagen and McLaren Automotive, among others. The car program cost, on average, roughly $1 billion annually (or nearly a fifth of Apple’s research and development budget a decade ago), with outside teams for chips, camera sensors, cloud services and software adding hundreds of millions of dollars more to the yearly spend.
But Apple never got close to realizing its original vision, or any of its subsequent ones. It didn’t get as far as testing a full-scale prototype on public roads. That it didn’t is partly thanks to the enormous technical difficulty of its self-driving goals, as well as the punishing economics of the automaking business. The project was also a failure, at the highest levels of the company, to settle on one thing and do it.
“There are a lot of roads you can take when you have a lot of really smart people and a very big budget,” says Reilly Brennan, a partner at the transportation technology venture fund Trucks VC. “But Apple never had the ability to make a bunch of specific decisions to lead them one way or the other.”
This story is based on conversations with several people involved in the development of the Apple car over the past decade, nearly all of whom asked to remain unnamed because the work was private. According to a longtime Apple executive who worked on the car, it was widely seen within the company as an ill-conceived product that needed to be put out of its misery. “The big arc was poor leadership that let the program linger, while everyone else in Apple was cringing,” they say. Asked what went wrong with the effort, a senior manager involved in the vehicle’s interior design replied: “What went right?”
What emerges is a portrait of the product development process at Apple today. The $2.6 trillion company has a history of hugely ambitious bets, and a track record of upending long-established businesses. It’s been a while since it did that, however. The iPhone is 17 years old and its sales declined last year, and newer products such as Apple’s watch and AirPods, while profitable, exist mostly in its orbit. The jury’s still out on the Vision Pro. Right now, the company is looking for its next big thing, and does not seem sure how to find it.
It was Steve Jobs who first floated the idea of a car at Apple. In the late 2000s, in a typically grand pronouncement, the company’s co-founder and CEO declared internally that Apple should have dominant technologies in all of the spaces in which people spend time: at home, at work and on the go. For many Americans, being in transit means being on the road, sometimes for hours a day. “We talked about what would be this generation’s new Volkswagen Beetle,” recalls Tony Fadell, who was the senior vice president of the iPod division under Jobs. In the wake of the 2008 financial crisis, with American car companies on the brink of failure, the Apple chief executive even floated the idea of acquiring General Motors Co. for pennies on the dollar.
That scheme was quickly abandoned, in part because Apple decided it would be a bad look and in part because of the need to focus on the iPhone. But in 2014, seeking a new multi-hundred-billion-dollar revenue stream, Cook began to focus again on cars. Apple executives weighing whether to enter the market joked with one another that they’d rather take on Detroit than a fellow tech giant: “Would you rather compete against Samsung or General Motors?” The profit margins in cars were far lower than in consumer electronics, but Apple was coming off a stretch during which it had reshaped not only the music industry but the mobile phone market. To its supporters, the idea of getting into automobiles had the potential to be, as one Apple executive puts it, “one more example of Apple entering a market very late and vanquishing it.” While the initial prototypes operated like traditional cars, these supporters eventually pursued more radical redesigns, invoking a transportation technology experience they said would “give people time back.” The ultimate plan was a living room on wheels where people who no longer needed to drive their cars could work or entertain themselves with Apple screens and services instead.
But before sketching out its own designs, Apple considered acquiring Tesla. At that point the electric-car maker’s success was far from assured, and its value was less than $30 billion, or a 20th of what it is today. Adrian Perica, Apple’s head of corporate development, held a series of meetings with Elon Musk. But Cook, who’d succeeded Jobs three years earlier, shut the deal down while negotiations were still at an early stage. Apple’s chief financial officer, Luca Maestri, formerly the General Motors CFO in Europe, argued that the car industry’s low margins were something the tech company couldn’t easily overcome.
Although the Tesla idea was abandoned, the ambitions didn’t go away. Apple’s newly minted hardware chief, Dan Riccio, received approval to start building a car engineering team, and he hired hundreds of engineers from the auto industry for what came to be known as Project Titan. The team working on the car was called the Special Projects Group. Within the company, it was difficult to find spare engineering talent, with attention focused on preparing for the upcoming Apple Watch release and, later, the iPhone X, but Riccio managed nonetheless to poach several dozen engineers from other projects. Early on, Jay Leno gave the team a tour of his garage for inspiration and a minor lesson in automotive history. Around this time, Riccio, rallying his troops, often would close with “Boys, let’s go build a car!”
The infighting began almost immediately. Maestri, the CFO, remained a skeptic, as did Craig Federighi, Apple’s software engineering chief, who had to donate personnel to what he considered a vanity project. Jony Ive, Apple’s design chief at the time, was more ambivalent, pushing for full driving autonomy but also expressing doubts about the wisdom of the endeavor. Some car fans on the Apple leadership team, including the company’s marketing executives, were resistant to building a product that didn’t look and feel like a car. Services head Eddy Cue suggested that it might be more prudent to just try to make a better Tesla rather than invent an entirely new category of machine.
Similar disagreements played out within Project Titan itself. Steve Zadesky, a former Ford engineer and iPhone executive in charge of much of the car effort, imagined the company starting off with limited self-driving features that could then be improved. Others held fast to Level 5. Perica, the mergers-and-acquisitions chief who’d pushed to buy Tesla, told the Apple car team that the company should build “the first bird,” not “the last dinosaur.” When the group first began staffing up in 2015, the goal was to bring something to market by 2020.
Under Ive, the microbus design emerged. The interior would be covered in stainless steel, wood and white fabric. Ive wanted to sell the car only in white and in a single configuration so it would be instantly recognizable, like the original iPod he’d designed. At one point, the group briefly discussed a more traditional SUV-like design, as well. The team’s secret facility in Sunnyvale, California, was packed with car cabin prototypes and simulators. “It looked like you entered Disneyland—it was chock-full of toys,” says someone who worked in the building.
The team played with several different ideas for the interior, including installing a pair of specialized touchscreens suspended with brackets from the ceiling to control the car from both sides of the cabin. It also engineered microphones to be placed outside the vehicle to bring external sounds into the cabin, something passengers in non-Apple cars did by rolling down a window. “They would add all sorts of crazy features to the car and then realize those were bad ideas and pull them back out, leading to another cockpit redesign,” says an Apple executive with knowledge of the frequent changes.
Throughout much of the process, Apple continued exploring partnerships. Riccio and Zadesky, years after Cook shot down buying Tesla, met with Musk to discuss ways they could collaborate, including the possibility of Tesla producing batteries for the Apple car. That prospect didn’t advance. Musk, who didn’t respond to a request for comment for this story, at one point publicly called Apple a “Tesla graveyard” full of engineers he’d fired. A few years later, he tried to restart talks with Cook as Tesla struggled to build the Model 3. Musk has said the Apple CEO wouldn’t meet.
Talks with Mercedes-Benz progressed further. For a few months, Apple and the German automaker actively worked on a partnership similar to the Tesla idea, but with a twist. Mercedes would manufacture Apple’s vehicle, while it would also sell its own cars with Apple’s self-driving platform and user interface. Apple eventually pulled out, in part because the early work gave its executives confidence they could build a car on their own, people involved in the failed deal say.
At other points, Apple held exploratory acquisition talks with car companies beyond Tesla. The closest it got to a deal was with McLaren. Some Apple executives believed that scooping up the British automaker, which makes a few thousand cars by hand each year and sells them to the super rich, would excite Jony Ive, who’d scaled back his involvement at Apple after the launch of the Apple Watch, and reengage him with the company. The proposed deal, before it fell apart, would have provided Ive with a new design studio in London. Other discussions with BMW AG and, much later, Canoo Inc.—an electric-vehicle startup with a decidedly Apple-esque design aesthetic—went nowhere.
By 2016, Apple hadn’t gotten far, and internal advocates of scaling back its car ambitions began to win out. After the board of directors and senior executives began questioning the program’s viability and asking pointed questions about its costs, there were discussions about shutting down the project. But then Riccio convinced Bob Mansfield, a legendary figure at the company for leading the hardware development of the original MacBook Air and iPad, to come out of semi-retirement to shake things up.
Mansfield was among the car skeptics at Apple. His task, as he saw it, was to find out what could be salvaged from the effort. After a few months of evaluation, he decided to focus more attention on the self-driving system than on a car itself. Autonomous software, he argued, could benefit Apple in other areas, even if the company never made an actual vehicle. Other executives, notably Perica, thought Apple could license such an AI system to other carmakers without dirtying its hands in the auto business itself. Over an 18-month period from 2016 to mid-2018, Apple laid off about 120 people, a significant portion of the car project’s head count, according to people with knowledge of the cuts.
Before Mansfield persuaded Field, the former Tesla executive, to take over for him, he and Cook did manage to agree on an interim direction for the company’s autonomous driving efforts: a self-driving shuttle made in collaboration with Volkswagen for Apple employees to use at its new headquarters in Cupertino, California. That project didn’t come to fruition, either. It was seen as a distraction, and Field shut it down. He also eventually shuttered Apple’s work on batteries and other components he felt Apple could just buy off the shelf.
Under Field, full autonomy continued to be a focus even as it grew to seem less attainable. The Arizona demonstration, which the team spent nine months preparing for, was essentially a proof of concept. The team tweaked the prototype software to take turns and curves slower than usual, to make extra sure it wouldn’t injure Cook. “It was well scripted and well laid out,” says someone involved in its creation. “The intent was to show Tim that if we built this product, this is what it would look like for the customer.” (Apple bought the Arizona test track outright a year after the demo.) The team spent a lot of time working on backup controls for such a car so that a driver could extract it from tricky road situations, such as a complex construction zone. The most fully developed steering wheel substitute looked like the controller that comes with an Xbox. “It should have been either all autonomy or a wheel and pedals,” one person involved in the car’s development says, adding that the company spent a lot of time working on ways to mitigate the issue rather than on the hard problem itself.
There were other, smaller dead ends over the course of the project. Apple started planning a multi-acre engineering campus in Silicon Valley where it would design cars, but never broke ground. At one point, Apple and Ford met to discuss a proposal from the American automotive giant to sell Apple cars from its Lincoln brand, an unglamorous make that’s well-represented in rental fleets, to test the self-driving system. The talks didn’t progress past an early meeting.
For Field, Mansfield and others on the team, Cook’s indecision was frustrating. “If Bob or Doug ever had a reasonable set of objectives, they could have shipped a car,” says someone who was deeply involved in the project. “They’d ask to take the next step, and Tim would frequently say, ‘Get me more data, and let me think about it.’” In that setting, it was hard to retain talent: engineers Apple hired for the project would grow convinced things weren’t going anywhere and find jobs elsewhere. After Mansfield retired, the company tried another leadership change to boost the self-driving system. It put Williams, the COO, and John Giannandrea, Apple’s machine learning chief, in charge of overseeing Field and the project. Field left a year later, in September 2021.
Under Lynch, Apple never got to a streetworthy prototype. The self-driving technology in the company’s fleet of customized Lexus SUVs did show enough promise that there were plans to expand it to more cities in late 2024, according to people with knowledge of the plan. The idea had always been to sell self-driving as a subscription service, as Tesla does with its driver-support features. Other paid add-ons, such as Apple Music and Apple TV+ streamed to the vehicle, would help make up for the uninspiring margins on car hardware. (Some internal estimates pegged Apple’s cost to produce the car at about $120,000, far more than the $85,000 the company had wanted to charge for it.)
Last year, Apple pivoted one last time. Designs were tweaked to move from Level 5 down to Level 2, the level of Tesla’s current Autopilot, which can control both speed and steering but is assistive technology for drivers rather than their replacement. In keeping with that, the new design also incorporated a more traditional automotive interface: a steering wheel and pedals. “They finally smartened up,” says an Apple executive. “I was like, ‘Guys, you could have done this 10 years ago!’”
But the company had ended up where it began a decade earlier, with a product little different from what was already on the market and a basic, not-great self-driving system. “Kevin Lynch is a sensible person,” says an Apple executive involved in the car decision making. “He tried to bring a pragmatic view to it.” When asked, he made clear that true autonomy might be another decade off. He seems to have finally convinced Apple’s leadership that that was a problem without an affordable or reliable solution in the foreseeable future.
Recently, members of the Apple car organization were studying how the company would produce the less-advanced car. It considered working with Magna International Inc., which builds some models for Jaguar, BMW and Mercedes. But the indecision at the top of the company filtered down, sapping morale. Apple declined to comment for this story, as did BMW, Ford, McLaren and VW. The former Apple executives named in this story didn’t respond to requests for comment. Neither did Canoo or Mercedes.
Around the beginning of 2024, Cook, who’s known for making decisions based on consensus, began seriously considering shutting down the project. “That’s when you started to see members of the leadership of Titan look for jobs at other companies and within Apple,” says someone with knowledge of the matter. People working on powertrains and other car-related engineering products began to depart.
On the evening of Monday, Feb. 26, the roughly 2,000 employees of Apple’s Special Projects Group received an email announcing a 10 a.m. all-hands meeting the next day. On Tuesday morning, the employees gathered in conference rooms and at desks at Apple’s Silicon Valley offices were told that Project Titan was winding down immediately. Lynch and Williams broke the news on a video call, and they didn’t explain the decision.
The meeting lasted about 12 minutes. Both men thanked the staffers for their work and got straight to the reorg and layoffs. Some employees would immediately get shifted to Apple’s AI division, and some would move over to software engineering. A chunk of the team, though, was immediately without a job. Hardware engineers would have the opportunity to apply for roles in other groups, but there aren’t spots for everyone. Other employees, such as the hundreds of car-specific engineers, test track technicians, self-driving car testers and automotive safety experts, received emails with their severance packages. As for the Arizona track, Apple is already working to sell it.
Original article by Mark Gurman and Drake Bennett from Bloomberg
