It wasn’t clear from the article, did they need physical access to the key fob? Feels like some details are missing and the raspberry pi angle is pointless.
"General purpose computer used to hack something" is not really a story. This attack requires a salvaged ECU from another Tesla, that's a more interesting and informative part of the story in my opinion. "Using a computer salvaged from a wrecked Tesla to pop the door on any model X" would be a much more interesting take!
>This kind of problem means it's time for a recall for any car I've ever owned, and I think that holds for the majority of the industry today, still.
And yet we aren't inundated with brand-wide OEM recalls due to these other cars being hacked, despite the fact that there are hundreds of times more of them sold than Teslas every year.
"Take the number of vehicles in the field, A, multiply by the probable rate of failure, B, multiply by the average out-of-court settlement, C. A times B times C equals X. If X is less than the cost of a recall, we don't do one."
Because, historically, such issues simply are not getting fixed.
Basically every keyfob sold before the mid 2000s got hacked via various methods. That's just a fact car manufacturers have not addressed retroactively.
Vulnerabilities were found in Ford's 2019 key fobs.
Let's not forget you can use a De Bruijn sequence of like 3200 numbers to unlock any Ford with Securicide. I don't know if Ford fixed it in newer models but as it was on 90% of vehicles they sold since the 90s, they opted to roll the dice and not recall.
My family owns a subaru and a honda. I've had to bring each of them in (the subaru, twice) for recalls that could have been fixed with OTA patches had the manufacturers been so prepared.
So I can't sign onto your "we aren't inundated" assertion.
Before I got rid of it in favor of the Subaru, my BMW had its key fob hacked, but the manufacturer opted not to recall it for a fix, in the US at least. I think similar models had gotten different treatment in Europe.
I could only speculate as to why the Tesla hack that's fixable OTA gets so much more attention than these others that apply to larger numbers of cars, but I have no inside knowledge and have done no study that would make my speculation on that front any more reliable than yours.
Whatever the reason is, it's not because the hacks aren't there, or because it takes something less than a recall to fix them. Why do you think it is?
in my experience bluetooth hacking just sucks on everything, with maybe python pybluez being the least sucky. I'd be interested to know what could be on the pi though, what is the current state of bluetooth tooling for research?
California's median house price is $706,900 [1] so a million-dollar home is only moderately above average.
Simply wait until the house is empty and break a window with a rock. No tools required.
Take 10 minutes to grab whatever valuables motivated you to rob the place, then get out of there. Even if they have a monitored alarm, it doesn't matter because you'll be gone before the response arrives.
Annoying title for the article. The point is that tesla cars are easy to break into, not that you need a raspberry pi to do so. But having a "raspberry pi" in the title gives you probably more views :|
Isn't the point that Tesla cars aren't easy to break into, you have to buy a ECU and use bluetooth, and it was fixed, whereas most cars are subject to a simple radio signal jam and replay, and are never fixed.
You need a salvaged controller from another Tesla, I'm not sure if that's "easy". It's also been patched already, per the article.
And in any case, the exploit here was that the door was opened, something that locksmiths have traditionally been offering as a paid service for decades with no computer hardware at all.
Really, the vulnerability angle isn't so interesting here at all. But it's a cool hack.
Adding the "Raspberry Pi" clickbait also sounds like a smear attempt to me. We already have enough of the "let's restrict general purpose computing because bad people" narrative.
That site is pure cancer. I opened the article in the same tab and then pressed back to come back to news.ycombinator.com, but it just showed me a bunch of blog spam from taboola which seemed to have hijacked the back button :-/
What Tesla brought to the game was insane performance for Electic cars, _and_ OTA updates for maps and software. And _just as importantly_ they actually update the software in meaningful ways.
Kia have none of these things.
Your statement is very similar to "Apple could slap their logo on any Chromebook and sell it for 2K!".
> What Tesla brought to the game was insane performance for Electic cars
I used to race electric cars in college. It was no secret even then that electric motors had insane performance. With a proportional throttle you had to learn early on to go easy or you would smoke your tires. Almost any electric car will smoke a ICE car, the first lap. Even the humble Nissan LEAF is under 6secs.
What Tesla really brought was capacity (we were still lead acid and NiMH at the time) and the Supercharger network. The battery tech and network are their largest asset. The cars themselves are pretty underwhelming design-wise.
First in line, I used to habitually accelerate from stoplights at full throttle, in my Leaf. This resulted in ICE vehicles screaming past me, after they finally detonated enough fossil fuel to catch up. I realized I was inflicting permanent psychological harm and bought an i3 from Manheim using borrowed dealer credentials and slapped a skull decal on the back window. It is known and accepted that BMWs with skulls are fast. Nobody cares.
Why are all their competitors so far behind in efficiency (Wh/mile or your preferred units), then?
Porsche Taycan is rated somewhere in the high 400s for Wh/mile while the Model S is rated under 300 Wh/mile. The Nissan Leaf is also about 300 Wh/mile with nowhere near the performance of the Model 3 which comes in at 250 Wh/mile.
Either Tesla has a substantial technology advantage or all of their competitors are just incredibly bad.
>Why are all their competitors so far behind in efficiency (Wh/mile or your preferred units), then?
The competition is behind, but not "so far behind".
A quick Google and you'll be able to find real world testing of these EVs, and as far as I can tell it's known that Tesla exaggerates the range of their cars, in some cases drastically.
That's not to say Tesla's tech isn't great (it is). But there are tradeoffs. The Taycan is ahead of the Model S in a few ways, too.
A single data point, but my Model X (2018 75D) has averaged 326 Wh/mile over the last 2.5 years, vs a rated efficiency of 315 Wh/mile. It was at 321 before covid lockdowns started, which shifted my driving habits toward more less-efficient car trips.
Based on that I have a hard time believing that Tesla substantially exaggerates range. I reckon my single data point is probably worth at least as much as autocar.co.uk's single data point based on a couple orders of magnitude less driving.
Also a single data point, but my Leaf averages a bit more than 4 miles per kWh, or slightly less than 250 Wh per mile. I'm not trying to make a claim that the Leaf is better--it depends greatly on the sort of driving that you are doing, but I don't buy that the Tesla has a significant advantage either.
Leaf is a compact hatchback, whereas Model X is an SUV. A better comparison would be Leaf and Model S, a sedan, which is rated at 241Wh/mi (and can reach this in actual use[0].)
Perhaps, I wasn't clear, but I was comparing to the 300 Wh per mile rating claimed for the Leaf upthread (I haven't looked it up for myself). I am comfortably beating that in real world use.
Edit: the point is that it is not clear that Tesla is much more efficient which was the original claim at the start of all this.
>What Tesla brought to the game was insane performance for Electic cars
Their software is cool, not sure they pioneered anything in "maps".
But as far as performance, I'm not sure that's true. Electric cars by their nature are fast. People have been putting electric setups in drag cars for a long time and getting ridiculously fast quarter mile times. There are entire racing leagues based around electric drive trains. From what I've read, the Porsche Taycan has better performance than any Tesla.
I think pwagland is referring to when Tesla launched their first car.
In the 1990s there were some CARB-motivated EVs that might do 100 miles and 70mph. Not exactly performance to write home about, and they had all been discontinued (and mostly destroyed) by 2008 when Tesla started selling the Roadster.
Essentially all you could get were golf carts, mobility scooters, and the G-Wiz.
So the idea that "electric cars by their nature are fast" was very much not in evidence at that time. The Tesla Roadster, with 200+ miles of range and a 125mph top speed, was one of the first demonstrations that an electric car didn't have to mean compromising on performance.
Heck I've even seen GM put their logo on the Nissan Van.
Also "simply not True" is you being incorrect. OTA always existed, it's just EXTREMELY expensive to send OTA. Tesla only has a few vehicles so it's cheap compared to Ford. And they take today's Stockholders money to pay for those updates.
Why are OTA updates so expensive? The price isn't prohibitive in other fields. I don't see how adding more cars would make that process more expensive, as if anything I'd expect a lot of the cost to come from QA and with more cars that cost gets spread out better.
I'd hazard life critical systems requiring a greater degree of assurance that the software update actually arrived at it's destination. Maintaining connectivity to a great number of network end nodes is not a trivial endeavor, even before you get into them being able to move around.
But who am I kidding. Someone probably just pays pays for a SIM to embed in an ECU with remote wake-up to phone home and grab an update, and the cost comes from all the telco networks that need to be contracted with to forward the traffic back home.
Tesla at least managed to release an infotainment system with a user experience you'd expect from modern smartphones, something many other car manufacturers still struggle with to this day (the navigation UI lagging is a standard example, even though we've had the processing power to make these smooth for over a decade).
Infotainment doesn't matter if it has no features. Where is android auto or apple carplay? Where is a HUD or true 360 camera? High trim Avalon's have all of this, and so does a PHEV rav4 primes which is cheaper and has far more utility then a tesla. Tesla is not actually ahead of anyone one else except in electric power train, "self driving", and battery tech. They're losing the self driving advantages every day to the other manufacturers. Tesla won't be on top forever unless they fix these problems.
And who can forget about it's famously bad build quality? Theyre lucky that electric doesn't have many of the issues that ICE cars do because tesla build quality is terrible and they get away with it due to the simple power train...
Couldn’t agree more. I’ve played with Tesla infotainment and I’d much much rather my CarPlay system and physical climate controls. I do want an EV and love the drivetrain of the Tesla’s but that’s the only part of them at all that seems to be better than any other modern premium car.
You're asserting that it's extremely buggy yet offer no evidence other than to Google it. When you Google "bmw infotainment issues", "gm infotainment issues", and "ford infotainment issues" you receive the same pages of results.
GM is currently facing a class action suit for the infotainment system in 2019-2020 vehicles. Ford settled a class action lawsuit for its MyTouch Infotainment system and Consumer Reports failed to recommend the new F-150 due to the Sync3 Infotainment system. Subaru also has a settled class action suit for its infotainment system. BMW iDrive is pretty much universally reviled.
The Tesla Infotainment system, warts and all, is considered one of if not the best.
Not to say Tesla doesn't have these issues, they probably do. Thing is if you go on Google and type "X issues" you will almost always find what you're looking for.
The researchers hacked several different other car manufacturers too. Tesla was the only one to react by releasing a security fix, the others simply threatened to sue.
> Tesla could put their logo on a Kia and it would sell to this demographic.
If it had
- fully electric propulsion
- a large charging station network
- cool styling
- an actually usable infotainment system
- OTA updates for everything
- knowledge you're funding R&D for new, more affordable battery systems
... maybe.
I didn't list advanced cruise control because many car makers have it. You could attribute the feeling of contributing to society's progress and the good feeling of buying from a company that actually cares about EVs to marketing if you're cynical (I wouldn't).
