Apple home robots wouldn't exist without the abandoned Apple Car
The $10 billion spent on research for the Apple Car wasn't lit on fire, and every rumor about the company's robotics program points to lessons learned and technology invented from the program.
Maybe some day Apple robots will drive Apple Cars -- image credit: Apple for the iMac G4
Invariably, when Apple develops something new, it gets introduced in a high-end product, and then trickles down over time. Think of how camera features used to always originate in the iPhone Pro line, and the next years come to the regular iPhone.
Now the same thing is happening all over again, except this time that high-end product never happened. It was the Apple Car, abandoned in 2024 after ten years and -- at least -- $10 billion of research and development.
Apple didn't throw that $10 billion away on a whim. It's not just gone, with nothing to show for it.
We're going to see the research for the Apple Car. We've already seen some. It just won't come in a moving vehicle.
It will come in CarPlay, obviously. It has already come in Apple Intelligence.
And, it will be seen the most in Apple's robotics efforts.
The repeated claim is that Apple will produce an iPad that goes on a robot arm. That iPad may or may not display some incredibly irritating emoji smiling face, but it will turn to follow you around the room.
That could equally be irritating, but there are clear accessibility benefits. Plus Apple will design it well, and if it weren't possible to make such a thing appealing, we wouldn't remember the Pixar lamp, or Apple's iMac G4.
Apple Car was never just about a car
It's become a convenient fiction to say that Apple is lagging behind the rest of the AI industry's research, and Apple itself hasn't always helped its case. But if anyone really believes Craig Federighi choked on his lunch on seeing ChatGPT in 2024, take a quick trip back to seven years earlier.
In 2017, Tim Cook finally caved in and revealed that, yes, Apple was looking at a car -- but not only a car.
"We're focusing on autonomous systems," he said. "And clearly, one purpose of autonomous systems are self-driving cars. There are others."
There are others. It's right there.
You know an Apple Car would have looked great.
"We sort of see it as the mother of all AI projects," he continued. "It's probably one of the most difficult AI projects actually to work on and so autonomy is something that's incredibly exciting for us, but we'll see where it takes us."
Just to absolutely nail down that Apple was using the car project to develop AI, Cook even added that, "we are being straightforward that it's a core technology that we view as very important."
Even Apple employees are said to have been glad that the Apple Car project was abandoned -- and partly because it meant they could move on to other AI development. But in every case, they would then bring with them what the Apple Car project had required them to learn.
What the Apple Car taught Apple
For a start, cars need absolutely flawless, uninterruptible, and fast environment sensing. Apple already gave HomePods the ability to alter their output based on what they detected in their surroundings, but that wasn't quite the same as the life-and-death situation that is hurtling down the highway at 70 miles per hour.
Then much later, iPads got Center Stage, the ability to automatically pan and re-frame cameras if the user moves, or if another user joins them.
That had to be done without moving the iPad, but a robot could detect the same movement and twist to better compensate. An iPad on a robot arm could turn around completely when it registers its owner coming up behind it.
Or a smart car would have to know how many people were in it. The loading of the car makes a difference to its handling, and there's also the far more minor issue of multiple people potentially arguing over which playlist to listen to.
Apple Car would presumably have had to deal with distinguishing between voices, and perhaps prioritizing them. In the car, maybe a system could determine that whoever is in the driving seat is the one to listen to.
An iPad can already move its lenses enough to offer Center Stage, but one on a robot arm could move much more freely -- image credit: Apple
But before you can weigh up whose voice to respond to, you have to parse all of the voices. HomePods and Siri already do some of this, although there is still a tendency for the wrong device to respond to a request.
With that improved, an iPad on a robot arm could at least swivel to face whoever was asking it something.
Then while all of the rumors seem to point to there being an iPad on a moveable arm, it isn't clear if the base would also move. If it does, the device would have to find its way around a room, avoiding bumping into people, and perhaps understanding what a user means when it tells the device to go help little Tommy with his homework in the den.
We do already have robot vacuum cleaners that bump into walls and criss-cross our floors. But the kind of navigation and environment sensing technology developed for an Apple Car, would make those vacuum cleaners seem like toys.
And Apple has robot tools already. Specifically, it has ones such as "Daisy," which are used to strip old iPhones for materials to recycle.
If that seems like a different development than the Apple Car, Tim Cook noted that Apple's recycling already depends on AI.
There is a difference between single-purpose tools like "Daisy," and more autonomous robots. But in May 2025, Apple revealed it was using human instructors wearing Apple Vision Pro, to train robots in various functions.
The future is already here
For Apple's own reasons, it decided to not pursue the Apple Car. Perhaps one day, there will be poorly researched books written about it, like there was about Steve Jobs after he died.
As a research bed, it has unequivocally been invaluable for Apple. And even before iPad robots, we're seeing benefits from the research done on it.
Back in 2022, for instance, an ex-Apple Car engineer revealed he'd worked on solving car motion sickness. Flash forward to 2024, and Apple launched Vehicle Motion Cues to prevent this for iPhone users.
That feature came directly from Apple Car research -- and it has already also made its way into Apple Vision Pro development.
Read on AppleInsider
Comments
Imagine if Apple had spent $10-20 billion and a decade on AI instead. Or if Apple recently had spent $200 million on retaining talent - like 0.7% of R&D or - what - 7% of legal. Or if Apple had 30% annual growth on data centers like MS instead of 6-7%. None of it happened.
Instead we're left to discuss the colors and lens positioning on the next iPhone as that is all we have due to a decade of failed projects and strategic blunders from Cook and his team. The markets have spoken. Jobs made Apple #1. Now it is #3.
I wish they simply scaled it back rather than cancelling it altogether. Cancelling it simply fanned the flames of those that thought it was a mistake to begin with, and those people usually can't see the bigger picture. We don't listen to them.
Apple was decades *ahead* of other companies in the AI field. The very first iPhone in 2007 featured a machine-learning powered on-screen keyboard, and their machine learning goes back further than that.
What took the world by storm was a *segment* of AI called "generative" based on Large Language Models — and YES, Apple missed the boat on that one with regards to Siri, and only Siri. Apple still sells the hardware that people use to consume generative AI from all of the competitors, and isn't computer hardware their primary focus? Why do they need to compete on every front?
They may not be an early adopter of generative AI, but given the misfires we are seeing from the likes of OpenAI, they were wise not to rush into it too quickly, as well.