I've read those, and while it is "theoretically possible" it's not practically speaking. All the hacking demonstrations have been done with specially prepped cars, often after wiring in debug hardware in order to read the ECU code beforehand.
Basically, anything is hackable. Anything. The question is how much work is it going to take to hack it? So the challenge is to harden your device so that it's not worth hacking. If I can hack an amazon locker to steal some packages, but it's gonna take me $50k to do it... I might as well just buy the amazon packages and save the time and money.
If you're gonna sneak into someone's garage, break into their car, and install something you might as well do it once, install some override hardware or a bomb, and be done with it. If you don't want to leave evidence just set it up to self immolate once it's done the dirty work. It's not worth sneaking in multiple times and doing a month of reverse-engineering just to hack the car.
You can get the same results for $1k for a super-fancy bomb, or $50k and a month on top level embedded engineering work and hijacking the satellites. Plus, imagine you get a month in to developing the malware and the target buys a new car? All your coding is now trash unless you really hate the person who bought the car. But the carbomb would be fully compatible with the vehicle.
I've read those, and while it is "theoretically possible" it's not practically speaking. All the hacking demonstrations have been done with specially prepped cars, often after wiring in debug hardware in order to read the ECU code beforehand.
Basically, anything is hackable. Anything. The question is how much work is it going to take to hack it? So the challenge is to harden your device so that it's not worth hacking. If I can hack an amazon locker to steal some packages, but it's gonna take me $50k to do it... I might as well just buy the amazon packages and save the time and money.
If you're gonna sneak into someone's garage, break into their car, and install something you might as well do it once, install some override hardware or a bomb, and be done with it. If you don't want to leave evidence just set it up to self immolate once it's done the dirty work. It's not worth sneaking in multiple times and doing a month of reverse-engineering just to hack the car.
You can get the same results for $1k for a super-fancy bomb, or $50k and a month on top level embedded engineering work and hijacking the satellites. Plus, imagine you get a month in to developing the malware and the target buys a new car? All your coding is now trash unless you really hate the person who bought the car. But the carbomb would be fully compatible with the vehicle.