Senator Wants Cybersecurity Answers from Automakers

[Guest]

Sen. Ed Markey, D-Mass., asks 20 major manufacturers tough questions about automotive security and driver privacy.

Senator Wants Cybersecurity Answers from Automakers : Read more

 

[ericburnby]

Complete waste of time and taxpayers money.

This is the field I work in (automotive control systems). The only reason they could "hack" into cars was because they had physical access to them and completely ripped the interior apart to get to the wiring and modules.

In the real world there's no way someone is going to wirelessly access your car and make changes to how things operate.

It would be like claiming you can break into my PC at home through the Internet when I left the Ethernet cable unplugged.

 

[AndrewJacksonZA]

@ericburnby: Including the bluetooth entertainment stuff? And what about cars that are wifi hotspots (admittedly I don't know of any cars that have wifi hotspots built in?)

 

[ddpruitt]

I have to agree that this is very unrealistic. Of course you can hack into a car if you took the dash apart. The other hacks use the OBD interface to reflash the firmware, something that's required. This is fairly easy to do because (wait for it) it's mandated by government regulations! Other than reflashing the ECU or changing the firmware on the radio, both of which require physical access, there isn't anything you do to hack a car.

 

[unksol]

It would be like claiming someone could break into your computer at home that was connected to the internet by WIRELESS. Which they CAN. Was the DARPA study a waste? Yes. Everyone knows you can access a vehicles systems via the CAN bus that's what its there for. That's why automakers dismissed that pointless study.

You're ignoring wifi, text messaging, remote access, internet access, blue tooth access and smart phone access that are being built into cars. And the level of system access or how far they are tied in. all of those can be hacked wirelessly if not properly secured.

I suppose you also didn't read the link about hackers being able to unlock, start, and locate a vehicle via GPS with text messages? and what handles those commands? The ECU. which runs code. If you have access to the ECU wirelessly it can and will be hacked.

 

[Darkk]

Just make sure you don't install McAfee anti-virus in your car otherwise it'll run real slow.

 

[derekullo]

"I suppose you also didn't read the link about hackers being able to unlock, start, and locate a vehicle via GPS with text messages? and what handles those commands? The ECU. which runs code. If you have access to the ECU wirelessly it can and will be hacked. "

http/www.technewsdaily.com/7932-high-tech-car-theft.html

"The thief attached a secret GPS tracker to the vehicle — and then locked it and left it where it was."

I did read that part lol.
They are using their own GPS tracking device not the GPS of the car.
So they would still need physical access to place their own GPS device.

 

[Hax0r778]

You don't need physical access to the car....

http/www.autosec.org/pubs/cars-oakland2010.pdf

http/www.autosec.org/pubs/cars-usenixsec2011.pdf

 

[TheSource49]

The relationship between the private sector and federal government in terms of cyber security will be very interesting to watch over the next few years. I think you would be hard pressed to argue that the government will not have some direction intervention into cyber security programs or at least the auditing of them.

Best practices in this field are often hard to identify, I would encourage you to read how companies like OPSWAT are advocating the use of multi-scanning and the higher detection rates that accompanies such a method

 

[TheSource49]

The relationship between the private sector and federal government in terms of cyber security will be very interesting to watch over the next few years. I think you would be hard pressed to argue that the government will not have some direction intervention into cyber security programs or at least the auditing of them.

Best practices in this field are often hard to identify, I would encourage you to read how companies like OPSWAT are advocating the use of multi-scanning and the higher detection rates that accompanies such a method

 

[cicilyqyo412]

if you need a job try this site JOBS61 (dot)¢øm. Dan does it at home and makes $25.98 hourly just sitting and typing stuff all day...No experience needed too

 

[f-14]

ericburnby
"...It would be like claiming you can break into my PC at home through the Internet when I left the Ethernet cable unplugged."

including your blue tooth wireless mouth, keyboard or headset or your wireless card? and if it's that laptop with wireless built right into it?

the study isn't a complete waste it's the start of a measurement metric that will become more important down the road as more wireless access and computer components gets built into automobiles.
nobody conceived of the idea of having computer parts in automobiles in the 60's and then not even a decade later EFI was introduced and ECM's were becoming standard in the 80's.

p.s. the wire harness doesn't have to be accessed thru the dash, the whole bundle goes right out the fire wall into the engine, which has no panel protection from underneath and hoods/bonnets on automobiles are easy for any thief to pop open to gain direct access from above.

plus i'm sure there is always a way for someone to hack into OnStar at the service centers themselves data bases full of names, addresses, phone numbers, gps information. then what do you do?

common criminals break a window, professionals work as a team and kick in every door as well as come in thru the windows. that's how anonymous works, if you say and think it doesn't happen.

 

[ericburnby]

To the people who mentioned wireless, let me explain how vehicles are set up.

CAN bus only one bus and most vehicles have more than one bus depending on what's being controlled. Further, vehicles can have more than one CAN bus. A high speed one dedicated to powertrain (engine, transmission, ABS) and a low speed one for other modules.

Entertainment & communication functions would have their own bus which would connect to the rest of the vehicle through a gateway which basically passes messages back and forth. In this way, a failure of a general module would not disrupt communications between safety related modules on a different bus. They communicate by passing messages.

There is no method in place whereby you could alter critical systems (like the engine) via wireless like Bluetooth or cellular EXCEPT for specific functions the automaker has implemented (for example, Onstar being able to turn your engine off if the police request them to). In this case the ECU will look for a specific message to perform the shutdown.

If automakers made it possible to actually reprogram vehicle modules through wireless or over cellular then I could see a problem. But this is a LONG way off. Programming vehicles is something automakers want dealers to do with the vehicle connected to a stable power supply ( charger) so programming can complete without incident.

Like I said, I work in this field (designing those very modules used in cars). If you want to know more just ask.

 

[ddpruitt]

You don't need physical access to the car....

http/www.autosec.org/pubs/cars-oakland2010.pdf

http/www.autosec.org/pubs/cars-usenixsec2011.pdf

Read more than just the abstracts. They needed physical access to the car to affect the drivetrain. All the wireless systems could affect where the convenience features. If you read between the lines you can see they tried to affect the drivetrain but couldn't. On top of that the experiments had some fairly sever flaws in them.

 

[stefansavage]

Hmm... if you actually read the second autosec paper, you'd find otherwise. I'll quote from it: "To be clear, for every vulnerability we demonstrate, we are able to obtain complete control over the vehicle’s systems." and then the paper goes on to describe this for several different channels including bluetooth and cellular. The claim that this is not possible is trotted out again and again, but is not backed up by the demonstrations described in this paper. As for the drivetrain comment, they were able to demonstrate turning the engine on and off and turning the brakes on and off. What they did not demonstrate was increasing acceleration. However, I think what is shown is enough to demonstrate safety concern. If there is a question about flaws, you should probably state what they are.