Well, I'm kind of hyped about autonomous cars and there are a bunch of advances right now. But how do you protect such technology? It's not like my pc will explode or hit shit (with me tied to it) if I get some malware on it.
Do you have any idea on how people are mitigating this?

I have no idea but its an interesting topic. If you look at flights recorders they are capable of being subjected to enormous forces so I suspect some of the protection technology could be repurposed.

@anth12 I was talking about directed malware, not environmental bugs. And in addition I am pretty sure airplane systems aren't connected to the Internet, cars will probably be and also probably communicate with surrounding cars (worms?)

Interesting question. Right now because it's not mianstream tech I don't suppose a lot of people are thinking about it. However I did see this video once about a Tesla being hacked (and controllable) remotely, which was interesting and kinda scary if you think about it.

How do you keep people from popping your tires or putting sugar in the gas tank or syphoning out your gas?

@aceface Doing that unseen is harder than hacking something, you can't use a VPN when syphoning gas ;) Also hacks allow for more sophisticated things, e.g. assassinations which look like accidents.

Well from a technical standpoint it can be a self contained readonly embedded system along with a full lock out with iptables, at this point you have p2p or central servers. p2p would have the problem of having to trust all communication between cars which could be faked by a hacker to create rush hour traffic when they want or maybe even beating the signal to the car first to make a car wreck. using a central server holds the risk of overload or a ddos attack, but if you store the fingerprint of the ssl cert this could be trusted more over p2p provided every car has a private key that is registered on the vehicles creation and wrecked cars are pulled from the database. this would prevent just anybody being able to do a mitm attack or beable to supply false information provided the private key in the firmware is protected by something like a gpg protected key using multiple memory chips

@aceface you can't spread a massive attack physically easily. Think about creating accidents in a city scale.

@h3ll with it not being read only it opens the air for rooting and fake ota updates. the only fully secure way I can think of is by what I said earlier with regenerated and revoked gpg key every flash update