Apple's Titan(ic) iCar project is dead as self-driving dream fails to materialize After nearly a decade of work, two indictments, the departure of a senior exec, and unknown levels of expenditure, Apple has reportedly decided to cancel its not-so-secret self-driving car effort, Project Titan. Apple hasn't publicly confirmed the decision – we've asked – but unnamed Cupertino insiders leaked news of the …
That project never made much sense for a company like Apple. Now, if they wanted to build the software stack for self-driving cars, that would be a whole other story. It'd be a bit outside their wheelhouse, but not as far as potentially getting into automotive manufacturing. They already have their own chip design facilities, they already have an infotainment system for cars, so they already have a couple of the big pieces needed to be a one-stop shop for the likes of Toyota, Honda, Nissan, Ford, etc. Granted the missing piece is probably the biggest and hardest to implement, but they do have the advantage where they could potentially design their own custom chips around the needs of the project. I just can't fathom why anyone would think it was a good idea to get into the automotive space as another manufacturer.
I think they just got conned into following Google in doing something that's way outside their domain. I actually think it makes more sense for Apple to build a self-driving car rather than just doing software, because Apple is primarily a hardware company, not a software company. If they had created something high-quality they could have convinced people to pay double usual prices. By the time they decided to just do the software I think the project was already dead. And when they saw Tesla having to cut prices to fight off competition that was probably the last nail in the coffin.
Admittedly it does not technically make a lot of sense for Google to do self-driving cars either; that was the old Google that would spend billions on a white elephant project just because it sounded cool (They literally had the motto "do cool things that matter" for a while). The current Google is concentrating on raising its stock price, and would never start such a project. I'm actually wondering how long they will keep it around, considering that it's very unlikely to be profitable for a long time, much less reach the kind of profit margins that are considered successful within Google.
Back in the days, you could be drunk on home-made hard Apple cider in the fields, and just let the horses drive you back home on their own, no sweat (bio-autonomous, way beyond ADAS)! Apple should indeed focus on this Amish horse buggy (truly smart transportation) and work-from-home stance (made possible by their very tech), IMHO, and develop a range of tasty (yet fashionable) ciders for their corporate bottom line. The more time people spend in their neighborhoods, including homestead, sidewalks, and local businesses, the better the economy (less vertical concentration of wealth into fewer and fewer, way overly-rich, hands).
Apple have all the money, close enough.
They can afford to try pretty much anything and sink absolutely massive amounts into it, and it will cost them almost nothing at all because it simply reduces their tax bill.
I expect that they have several pie-in-the-sky projects that aren't expected to make any revenue but might possibly do so eventually, and it only costs them cashflow as they can write it all off against tax.
"I expect that they have several pie-in-the-sky projects that aren't expected to make any revenue"
Remember, this is Apple. They don't want the revenue, they'll just sit on the large number of patents they've been granted for spurious ideas and concepts from these projects until they can sue another company for patent infringement after bringing a new product to market. It's what apple seems to be best at these days. They aren't known as a 'vexatious litigant' for nothing.
They can afford to try pretty much anything and sink absolutely massive amounts into it, and it will cost them almost nothing at all because it simply reduces their tax bill.
Richard 12,
US corporation tax is 21%. So if I don't spend $1 billion on a self-driving car division this year, I have now got $790 million dollars after tax to give to shareholders as dividends or put into my enormous cash-pile. This will also be true next year, and for every year after that I don't spend my $1bn on doing it.
If I do spend that billion - then I'm getting to offset that spending against my profits. So I've now reduced my profits by $1bn - which means I've offset $210m off my taxable profits - which means it's only cost me $790m to get a billion's worth of work done. Seeing as it hasn't worked, and they've been running it for nearly a decade that means they've got no billions worth of work done for about $8 billion spent.
I'm sure a billion is too high here, but the department did employ 2,000 people, so it's going to be in the hundreds of millions spent.
No chance Apple would have built software for others. What's the value add, unless they alone solve Level 5 automation? Otherwise the software is mostly invisible to the driver, who only interacts with the car to tell it "take us to grandma's house" and then everyone sticks their head in their phone or watches a movie on their personal display in front of them. What's the branding angle for Apple being just software in that car, totally hidden from the passengers? Why would an OEM pay $1 more per car to Apple for this versus a competitor who also markets Level 5 automation software?
The only way to make it work was for Apple to make a car. Or rather DESIGN a car, and presumably contract out the manufacturing. The problem is that while contracting out the manufacturing of small items that are basically 100% and 0% mechanical content like phones (other than the physical buttons, which Apple wants to eliminate) and laptops (the hinge and the keyboard being the only mechanical items there) the manufacture of a big item like a car which totally flips that with overwhelmingly more mechanical items even in a modern EV.
A bigger problem is that the "premium" auto segment may no longer exist when we finally crack Level 5 automation. If you are just going to sit in that car not doing the driving (and therefore not caring about handling, acceleration and other things people pay more for in today's cars) and your only interaction with the car is to tell it where to go and use the display in front of you like a dumb wireless monitor for your phone (or whatever replaces the smartphone in the future) then what's the value proposition of a car that costs several times as much as the average car? I mean heated leather seats versus cloth, OK, but I think the market for premium vehicles is going to crash when Level 5 EVs arrive. It will still exist, some people will pay more just to be seen, but the days where the kind of car you drive was part of your identity are already gone. Gen Z think the idea that people would care whether you have a Ford or a Chevy or a Mercedes is ridiculous, just ask them. That all gets even less important when you aren't even behind the wheel, because there is no wheel.
What's the value add, unless they alone solve Level 5 automation?
Software generally isn't the area of expertise for car makers, as evidenced by multiple utterly shitty infotainment systems. Software, however, is one of Apple's areas of expertise. Plus, as I said, Apple can be a one-stop solution. The chip to power the infotainment system AND autonomous driving system, plus the software to go with it. So, at least in theory, they can be a sort of plug-and-play solution. You buy the SoC and software from Apple, then you don't have to worry about the infotainment/autonomous driving aspect of things. You can just focus on the mechanical parts of the car, as a car maker.
Of course the better your software is at handling all the various things that can happen when driving, the more in demand your product will be. It might just be that it's finally sinking in just how hard autonomous driving really is. We spend literally thousands of hours practicing driving, and we still suck at it as humans. Getting a car to drive in a more or less straight line and obey speed limits is easy. What's hard is dealing with situations like a deer running out into the road, or do you prioritize the passenger(s) or pedestrians if the accident can't be avoided? It's all the various edge cases that you have to contend with that's holding things back.
But, you do raise an interesting sociological issue that will be interesting to watch play out. Assuming I'm alive long enough to see Lvl5 cars become commonplace, what will happen to all the people who love to buy performance cars? I live not too far from a mountainous area and on the weekends you always have people with their sports cars out driving along the winding roads for fun. Also seems like rarely a weekend goes by when you won't find the cops and fire department dragging a car up that went over the side and down a few hundred foot drop. People who like having sporty cars seem like they're in for a rough adjustment period.
Apple only sell phones to people using iPhone... it is working quite well. A lot of people buy things like TVs based on it supporting Airplay already - nobody needs access to 100% of the market to do well. And demographically, iPhone users are probably those with the money to spend on a premium-branded product like iCar.
Nio do it the other way round. They made the car first and now have a whole range of Nip products includning a smartphone. In fact NIo is problably the car company that is closest to creating an Apple like fan base, The have Nio homes Homes, Nio clothes collections, and so on https://www.nio.com/da_DK/nio-life.
The major difference between NIO's cars and Apples Phones is that in the cars, you can change the battery!
You could presumably force Apple to make a CarPlay app for Android to avoid regulatory probing. Or, presumably you could force Apple to make CarPlay an open spec, so anyone can implement it. If you're really hell bent on using Android Auto, you may be limited in your options for cars you can buy, but that's on you.
> Software generally isn't the area of expertise for car makers, as evidenced by multiple utterly shitty infotainment systems.
Oh it's worse than that. Anyone read this article in the Observer on sunday, discussing how Kia and Hyundai cars are being stolen in seconds by ne'er-do-wells wielding a device packaged to look like a gameboy?
I certainly did, because I have a Kia. And after a lot of digging I eventually found this paper from 2019. I'll sum up: cars communicate with electronic key fobs using a cipher with an 80-bit key, but of those 10 bytes Kia has hard-coded four of them and made three of the others derivable from the first three, reducing it to a 24-bit key. This can be brute forced in seconds, and all it requires is one communication from the key *searching* for a key - the key doesn't even have to be present.
So I for one would welcome the entire software and firmware stack to be taken from the cockwombles that made that decision. If you have a Kia or Hyundai I put a bit more on this reddit thread, including what I think is a mitigation.
Thought for the day: if you fuck up your encryption and your customer data is stolen, you have to report yourself to the information commissioner, contact your customers and eat humble pie. But if you fuck up your encryption and your cars are stolen, you don't have to do a damn thing. Because apparently, it's not your fault.
Assuming I'm alive long enough to see Lvl5 cars become commonplace, what will happen to all the people who love to buy performance cars?
They'll still buy performance cars, just as they do despite the existence of SUVs, automatic gearboxes and SUVs with automatic gearboxes.
Granted automotive infotainment systems suck, but look at how people are using those in their cars today - i.e. they aren't. They use CarPlay / Android Auto to interface with their phone. What people really want is for that capability to be extended so that the display is owned by your phone. If you have an iPhone it will have a CarPlay interface, if you have an Android it uses that interface.
Some automakers will be resistant to this because they no doubt hope to use that as a continuing source of revenue, but consumer demand will eventually win out. Any automaker that tries to force people to use their own will lose in the marketplace. Doubly so with autonomous cars, because in those even the owner is always a passenger rather than often being the driver who needs to devote most of their attention to that task (ideally all of it, but we all know that's not the case in the real world) In a Level 5 vehicle there will be a big display in front of each seat, and it'll interface with your phone. There won't be a role for a native infotainment system, because no one is going to buy a car based on that (unless it doesn't support carplay/AA well in which case people will NOT buy a car based on that!)
"People who like having sporty cars seem like they're in for a rough adjustment period."
Well, people who like driving will be in for a rough adjustment period if autonomous cars are all that is available. Even non-autonomous cars will have levels of automation that probably prevent fun driving.
Personally I think sports cars like that will survive as an enthusiast niche, much like horseriding now. There is no logical reason to ride a horse nowadays, but many people spend their weekends doing it. People will still drive (likely classic) cars for fun on country roads, but they will be enthusiasts.
I expect that by the time I'm so old the only way I am able to hit the road is because cars can drive themselves manual driving will be banned in most places except very rural roads. The kinds of places that allow ATVs on the roads today will allow human driven vehicles. On major or minor highways and streets in towns bigger than 5000 or so human driven vehicles will be banned.
I don't think you can get into the stack without access to all the mechanical gubbins, which is why Google was an early investor in driving cars around with cameras and other systems.
I suspect the end game may have been to get into some kind of premium "mobility service", think of how differently Apple would have run Uber. Having seen high margins failing to appear, they may have decided to get out. But they will probably be able to use some stuff to inflate their "Car Play" software which does turn a tidy profit thanks to all the fanbois keen to use it in their cans.
Who wants this stuff?
Whenever I hire a car all the latest look-at-me I-can-do-it myself electronic gizmos just make me ever more unlikely to buy one. It's bad enough when machines beep at you for doing something approved of by the Highway Code (taking off your seat belt in order to look behind when reversing), quite another for it to attempt to wrest control entirely.
For some reason Americans overwhelmingly favour automatic gearboxes. Europeans prefer to select gears themselves. I find the idea that we (Europeans) would choose to let software do the whole kerfuffle frankly risible. Cheapskate delivery companies, sure. Taxis, maybe.
Ah, taxis.
Being something which costs an enormous amount of money a car is a pretty irrational purchase. For most people it would make more financial sense to take taxis when public transport doesn't suffice. But that would mean sharing with other people! Communism!
One rationalisation is that personal transport means personal choice.
Of necessity autonous cars are not about personal choice. You can't argue with one about how if it doesn't go a little bit faster than is strictly allowed then you will miss your plane.
Might as well get a taxi.
-A.
If your shopping isn't strapped in it will become a projectile the event of a crash just like a passenger. Or throw itself over you, risk a can of beans under the brake pedal etc. It is a legitimate safety issue, speaking as someone who has done this ;)
Cars have these facilities called boots and footwells for storing things.
I recently upgraded to a car with more smarts under the bonnet. It randomly beeps at me: But none of the high-tech screens give me a clue as to why it's beeping at me.
I also find the driving "style" it wants me to adopt, whilst the ideal, is rarely practical. I often have to tturn off some of those clever features so the car is driveable.
I recently upgraded to a car with more smarts under the bonnet. It randomly beeps at me: But none of the high-tech screens give me a clue as to why it's beeping at me.
My other half has a Fiat 500 which puts the engine management light on for everything from "one of the sensors gave an odd reading for two seconds but it's fine now" to "stop driving RIGHT NOW" and gives no indications whatsoever what level of severity is involved.
'I often have to tturn off some of those clever features so the car is driveable.'
Be thankful you still have that option. I had a golf as a hire car the other day and it was impossible to turn off any of the incessant beeping/warnings. Then it locked up the brakes and wouldn't let me move the car because a few strands of long grass blew across the front and rear of the car and triggered the radar...
I think the whole Europeans ‘prefer’ manual gearboxes is a little spurious. Automatic gearboxes have traditionally cost a significant chunk more than a manual and in a region that is a bit less car-centric than USA, the cost is significant. Add to this that the major of people have never tried automatic to compare, then it’s not a true preference. I thought I’d never want an automatic, but circumstances meant that I ended up with one, it’s now very unlikely that I would choose manual again. My other half won’t even try to drive my automatic for some irrational reason I don’t fully understand. Add to this that most people in the UK learn to drive a manual, means that people are more likely to stick with what they know. Learning and taking a test in an automatic precludes you from driving manual and is a less common option.
In Europe, generally, auto gearbox is optional extra - i.e. costs more.
In USA, generally, manual box is optional extra - i.e. costs more.
Auto gearboxes these days are not the power hogs they used to be and are arguably more efficient in that they change gear in the "ideal" rev range.
I thought I’d never want an automatic, but circumstances meant that I ended up with one, it’s now very unlikely that I would choose manual again.
Shrug. If we're going to argue from subjective anecdote: I learned to drive with an automatic, learned to drive a manual a couple of years later, have owned several of each and driven quite a few more besides, and always prefer a manual. And I'm not a typical petrolhead; "performance" cars bore me and I've very rarely driven simply for the pleasure of driving.
I learned to drive manual, did it that way for 20 years but eventually switched to automatic after owning a manual Saab with a heavy clutch that was tiring to drive in heavy traffic. Manual gearboxes are not an option if you are driving hybrid (which I am these days) or electric. I don't miss manual gearboxes and when I hired a van with manual gears recently spent an upsetting three minutes trying to make the f***ing thing go into reverse as every manufacturer does that just differently enough to make it confusing. In this case the trick was liffting up a neat little ring partway down the gearstick.
Automatic gearboxes used to be pretty inefficient and unsuited to lots of gear changes. Less of a problem in the US where petrol is cheap and most journeys involve few gear changes, more so in Europe's crowded cities and their narrow streets. Nowadays automatic gearboxes are probably better than most people at shifting and a prerequisite for safe-driving systems like tempomatics and distance keepers. In fact, they'd probably be standard if they weren't a way to charge a premium: Skoda used to offer them on its cheaper models but VW made them drop it.
I recently had a problem with a car-hire booking and ended up being given a very fancy Mercedes for the week instead of the bottom of the range Ford (or whatever) I'd asked for. This thing would sometimes brake all by itself. The wing mirror flashed distractingly when a car I'd seen ages before was about to overtake. The steering wheel rumbled if I changed lanes to overtake without indicating (ok) and when I pulled back into the slower lane without indicating (not ok, according to my rozzer friends). Finally, it took three of us (three PhDs, two full Professorships) ten minutes to work out how to get the petrol filler open. It went like the clappers, but didn't feel fully connected to the ground even at quite legal speeds. I absolutely hated it. It did project a lovely Mercedes logo onto the ground when you opened the doors though. Time to stock up on vintage BMWs or something.
Which is why I'm sticking with my 1989 BMW E30 cabrio that I've owned for over 20 years.
When I first owned it I was living in central London so the automatic gearbox was a real benefit in that sort of traffic. And then I taught myself to left foot brake which is relatively straightforward with only two pedals, and that half-second quicker on the brakes definitely saved me from a massive coming together with a white van some years ago.
non-self-driving cars can quite easily (and sometimes do) prevent you exceeding speed limits, etc.
Automatic gearboxes are wonderful (mostly) this is not really something either side 'chooses' it is a generational cultural thing - you grow up with whichever one and you take for granted that's what everyone does.
As we move to EV and so on, gearboxes will soon be a thing of the past anyway.
A Taxi is a huge inconvenience if you do not live in an urban or semi-urban setting. The idea I could go to a friend's house for dinner, or to the pub in my nearest town, without worrying how to get back, is somewhat attractive as a rural person now that the culture/laws on drink-driving have (thankfully) tightened up so much.
Automatic gearboxes are wonderful (mostly) this is not really something either side 'chooses' it is a generational cultural thing - you grow up with whichever one and you take for granted that's what everyone does.
Again, this generalization doesn't hold for me. I know it's difficult for some to avoid assuming your beliefs are true for everyone; but they are not.
Being something which costs an enormous amount of money a car is a pretty irrational purchase. For most people it would make more financial sense to take taxis when public transport doesn't suffice. But that would mean sharing with other people! Communism!
As someone who can't drive due to vision issues (i.e. not by choice) I do save money on not owning a car. However that's partly reflected in me not being able to do things I'd like to using public transport - and so just missing out on doing stuff I'd rather pay to do.
Getting my bus to work at £4.50 a day is probably a smiliar cost to car ownership. But my journey might take anything between 25 and 40 minutes by bus - probably 10-20 by car and 35 if I walk. Which is good I suppose, I walk unless it's raining.
But visiting my brother who lives 30 minute's drive away in a small village is impossible by public transport and a half day's walk.
If I want to get somewhere on time I have to factor in that random buses just don't run, so you'd always need to use an earlier one. But if you order a taxi they might pick you up in ten minutes or they might be busy and be 40.
The Blood Transfusion Service phoned me on Monday wanting me to donate plasma, instead of the whole blood I currently do. But that means going to Reading. It's in the centre, so I guess a 50 minute drive, half hour if I pick the lowest traffic time. Or there's a regular bus that takes 105 minutes. So two hours to donate, 3 and a half to travel. A return is probably a bit less than the petrol cost (and there's no parking to pay) - so financially cheaper, but then how much is two hours of my time worth?
I'd say a car is a perfectly rational financial choice. Unless you live in a big city, with good public transport and lots of traffic. Your time is valuable, after all you charge your employers for their use of it. So even where public transport is cheaper, it's rarely as convenient. So even if there are available fleets of self-driving cars to hire, I'd image you'll still get people wanting to own their car - so they can use it whenever they feel the need.
Also my figures are for just me. If you're sharing journeys, cars get much cheaper than buses or trains. Obviously not taxis. But then journeys longer than 20-30 minutes start to get quite expensive by taxi.
Whenever I hire a car all the latest look-at-me I-can-do-it myself electronic gizmos just make me ever more unlikely to buy one.
I have a 2010 VW Beetle convertible, bought secondhand. Nice car, until the passenger door wouldn't unlock. Turned out to be a subtle wiring fault, but only after changing the release mechanism, the computer in the window-winder motor (sic) and the climate control computer (sic - it works the locks too) ... and all that after cutting the trim off the inside of the door because the very first stage in diagnosing "door will not unlock" in one of these things is "open the door".
My 1973 Citroen does the same job with a metal pull rod which is (a) almost failproof (b) easily accessible and (c) easily fixed if something does go wrong.
?For most people it would make more financial sense to take taxis when public transport doesn't suffice. But that would mean sharing with other people! Communism!?
People in the outer suburbs or rural areas would be staring down the barrel of a $100 taxi ride every time 'public transport doesn't suffice'
I assume you live in the inner city. Both because of the fact you seem unaware of the costs of taxis long distance, and because of the condescending way you sneer 'communism' at people whose circumstances you don't understand...
Face it - our roads have been cobbled together over the past century, mostly badly, with incremental improvements but in many instances left to fall apart. Throw an autonomous vehicle on to that lot - alongside masses of other vehicles wrangled by less-than-rational beings? Mad.
Now if Apple had been given the chance to re-build the entire California road system ...
And this is why creating a society based on moving people for hours a day for the most basic of human operations is doomed to always fail. Every single transport system on earth is a failure. Spending 10minutes or 2 hours commuting each way is dumb and an enormous waste of resources and time. People sitting on a train for 3 hours a day is not smart or productive.
It made sense for them to be in that space - they have custom silicon, some of the top industrial design and manufacturing experience and a presence in a lot of people's daily lives.
On top of that they have the problem of "what to spend the profits on". For a company the size of Apple, the only bets to make are big ones. They can't be seen as dabbling in small inconsequential markets (which is a shame because they could make a difference) - only stuff of global significance. Then it doesn't matter if the project succeeds or not, it's a tax write off anyway and there might be some useful patents and discoveries along the way.
That said, sustaining such a project needs an evangelist, which is subject to the usual rotation of senior management and depends on being able to demonstrate at least some progress. If, with all their skills in silicon, cameras and vision processing, they couldn't see the way (no pun intended) to some sort of product, then it's time to pull the plug.
Of course the question increasingly becomes: If others aren't even hanging on to see their competitors launch a product, is this evidence that level 5 autonomy is beyond our ability to produce? Even if you can't do it yourself, if you've got presence in the market and know that your competitor is going to launch something that works then it makes sense to hang on - you might be able to ride on the coat tails of their technical implementation and you'll almost certainly get the investor buzz of being in the right space. From this angle it certainly looks like Apple don't think Tesla (Waymo, Cruise and the others) are a serious threat, which is a bit of a problem for companies that want to be the "Apple of cars".
There are doubtless issues to be worked out in all these systems and it is a fiendishly difficult situation when you have unpredictable human drivers all over the place.
But, human drivers frequently drive into trucks, wet concrete, the sea (blindly following GPS), block roads, run over pedestrians and so many other things.
I doubt any driver reading this has never made some sort of daft mistake or very near miss at the hands of someone else.
If the argument is the program should be stopped because there are some accidents, then human drivers should not be allowed either because we have about 100 years of data showing they cause accidents, a LOT.
The question has to be do AI/automatic drivers cause fewer or more accidents. If the answer is substantially fewer (I have no idea) then those developing these systems should be encouraged and protected from the fear of litigation. If we look at medicine, it is known all surgery has risk but we still perform heart operations when the probability of success greatly outweighs failure. We wouldn't stop performing surgery because sometimes people die - though we do have strict protocols to investigate WHY each death occurred in case there is something wrong with the procedure (or those performing it)
JDX,
My suspicion is that if we were starting from scratch, autonomous cars might be possible now. If every other car was autonomous - and no pedestrians or cyclists were allowed on the roads. We've probably got the tech to do that.
But we aren't. And also lots of the self-driving tech is being trialled in places with relatively benign and simple road networks. And it's going to struggle even more when they try to do it somewhere where the roads were designed for horses and carts by people with an allergy to straight lines.
The problem is that computers don't understand anything. They just do what they're programmed for. Which is fine with aircraft autopilots. Which you operate in highly contolled conditions with highly trained human air traffic controllers and pilots around to deal with the edge-cases.
Roads aren't like that. And worse, the industry doing the the work is one that's famous for being led by visionaries who are often happy to release stuff that barely works - in the hope that they can keep fixing it later. The automotive and aviation industries are mature industries with tough standards that are heavily regulated and tend to be run by boring cautious management. The computer industry is barely regulated, immature, and often run by the lunatic who can best convince the Venture Capitalists that they can make them billions of dollars in just ten years. This is not a good mix, and in something safety-critical I think too many of the tech companies just aren't up to the task.
If every other car was autonomous - and no pedestrians or cyclists were allowed on the roads. We've probably got the tech to do that.
I really don't think we do. Let's consider some failure modes.
The "Top Gear Kevin Failure": If cars continue to be privately owned, some percentage will be owned by "a bloke named Kevin" who does his own maintenance with no idea what he's doing. In general, if maintenance isn't regulated and enforced (and certainly no US states are currently doing a decent job of this), you'll have a variety of nasty failures happening on the road, and other vehicles will have to account for them. I've seen cars lose parts (bumpers, for example) on the highway. I've seen a wheel come off a Jeep when the axle sheared. A friend was riding in a rear-wheel-drive car which had its driveshaft decouple from the torque converter, fall down, and hit a crack in the paving and burrow under it, bringing the vehicle to a dramatic stop as the rear axle gleefully disassembled. You probably don't want to be in an AV following behind that, because I'm betting its tires don't provide enough force of friction to decelerate to a stop in time.
The "Transponder MTBF Failure": People keep saying that AVs with transponders will let us do away with external traffic flow control (lights, stop signs, etc) and allow close-following convoys to reduce drag. Sure, up until someone's transponder fails.
The "Private Roadway Failure": I live on a private road. So do lots of people around here. Plenty regularly drive on Forest Service roads or powerline access roads. It's no small task for AVs to deal with these; they're often not in the mapping information, are often poorly maintained, typically aren't signed, and are generally gravel (or worse). So AVs will need a way to let humans take control, or we'll have to solve the last-mile (sometimes last-quite-a-few-miles) problem some other way; and if humans rarely drive, they likely will be even worse at it than they are now.
The "It's Not an AV Failure": Sure, forbid pedestrians, bicycles, etc. Animals get on roads. People get on roads even when they're not supposed to; how many times have you seen someone walking on the side of a limited-access highway, or a railway, or the like? Other roadway obstacles include things like rockslides and flooding, so the AVs will have to be able to continuously monitor for non-AV obstacles. And they'll also have to be able to distinguish non-obstacles. Trying to drive an AV though eastern Colorado (a trip I make several times a year) would be miserable if it kept slowing or stopping for tumbleweeds — and Colorado isn't even a prominent tumbleweed area, compared to some others. So AVs tracking one another might reduce the scale of the environmental-awareness problem, but it doesn't change it qualitatively.
The "Software Sucks Failure": Software sucks. Sorry, trufans, but that includes Apple software; you really don't want your AV stumbling into the next goto fail. Nontrivial software is really complicated, and complicated systems are difficult to get right and have revenge effects. And security failures in systems that lives depend on are kind of problematic, too.
Sure, it's easy to argue that if the public roads were all AVs rather than all human-controlled vehicles that the — admittedly terrible — accident rate would be lower, perhaps much lower. But there are still a host of failure cases that require both overlapping safety systems on the one hand, and the option for human takeover on the other.
Michael Wojcik,
I don't disagree with you. I don't think self-driving is happening any time soon, and its best chance is in very restricted circumstances.
On the other hand, there are also many faults with human drivers - and our current system allows a regular and large number of deaths every year. I think the testing that's been happening, in very benign environments, suggests that they're at least not hugely worse than human drivers. Even if they're bad in different ways.
All these problems combine to make self-driving a very hard sell.
Even my 2015 Volvo XC70 is over 320,000 Km, with little in the way of significant maintenance required. (Had to replace the starter, which fortunately is a pretty cheap repair on these; and the positive battery cable, which isn't particularly cheap, but isn't that bad either.) And it cost a bit more than a third of that estimated $100K.
I paid around $4K for my 1993 Toyota Truck, which has even more on the odometer. And that's a straightforward 4-cylindar engine in a relatively large compartment that's pretty easy to work on.
If I had stupid billionaire money to throw around, I can sort of see the appeal of buying a really expensive car — something obscure and fancy — just for the fun of it. But if I'm buying for practical reasons? No, I wouldn't be considering the hypothetical iCar either.
I'd always assumed this was a combination of PR, and that really rare thing, actual serious Apple research.
Not cancelled because it needs windows, but likely too much competition and probably was never a serious product development.
Telsa will have problems as traditional car makers produce more electric cars and stop buyying carbon credits from them.
iTV
* 42 inch LED TV
* aesthetically rounded corners
* available in 8 different shades of white
* requires Apple Vision Pro
* requires iPhone 16+ Max
* subscription only $89.99/month (iTunes content extra)
* supports apps (except Netflix, Spotify, Amazon Prime, paramount+, Disney+, Hulu, YouTube, BBC iPlayer, etc)
I'm not sure we ever will get full self driving. There is too much processing required for it to happen in car, and I'm not sure our various mobile networks will take the strain of hundreds of thousands of cars transferring potentially an awful lot of data per second. QoS won't help much as most of the data will be urgent, so have a high QoS priority..
Part of me would like to see it. I quite like the idea getting in my car, pushing a few buttons, then sitting there for a couple of hours as I am driven somewhere. But, I can also see it would be unsafe. For a few reasons:
1) Computers (even AIs) tend to work better when all the variables are predictable. IE, not really variables. People are not generally predictable. How is the AI driving your car going to cope when some person thinks it's funny to dash out into the road in front of your car?
2) Software, even AI based, is prone to bugs. Whether in the code itself, or in the training material given to the AI. What if a bug causes the AI to decide that the road on the other side of the crash barrier is where it wants to be?
3) Unknown situations. How does the AI deal with the unknown? I don't have a lot of experience with AIs, but from what I've seen, they tend to perform better in situations where there is a lot of data to back them up. One example I saw of an AI failing was in an article I read about the Law said that while one of the AI's (ChaptGPT I think) was good at passing the bar exams (because it had previous exams to go by), it did not do well when fielding general law queries, because it was dealing with situations not really covered by it's training data.
Full self-driving is still a long way off. Assisted driving can be a useful step. With an aging population there's a lot to be said for implementing features like collision avoidance more widely to make it safer for those with slower reaction times and perhaps other age and health related limitations to drive. Useful too for young inexperienced drivers. There are also issues with the overall "user experience" that need attention not just the in-car experience but the massive growth in road signage to the extent that in places there's just too much to take in and honest mistakes become inevitable. Signage should be being supplemented (and eventually largely phased out) by vehicle-readable digital "signage" that the vehicle can use to assist driving and navigation.
A few years back we were always being told that self-driving cars were just around the corner (ho ho). I was sceptical then because it looked to me like all the successful testing was being done in relatively simple environments, and nothing as complicated as say London or Paris. For one thing, in busy areas with a mix of cars and pedestrians, the humans in the process will negotiate the space by making eye contact and using other non-verbal cues. Machines can't do that yet.
For self-driving vehicles to be viable, you would have to further restrict freedoms for pedestrians, and manufacturers would lobby for exactly that.
In fact history would repeat itself: https://www.bbc.co.uk/news/magazine-26073797 . I drive, but I find the idea that people's freedom in urban environments should be curtailed for the convenience of the car lobby rather dystopian.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
