Skip to main content

Karamba’s Carwall thwarts mass hacks

Karamba Security’s Carwall software is said to prevent ‘mass hacks’ of vehicles’ on-board systems including those for connected and autonomous driving. Carwall sits in the vehicle ECUs and ‘learns’ the factory settings. If hackers breach the manufacturer’s cyber security and tries to infect the ECUs of in-service vehicles, Karamba’s software detects the impending change to factory settings and blocks activation. David Barzilai, the company’s chairman and co-founder, said with tens of millions of l
September 13, 2016 Read time: 2 mins
8519 Karamba Security’s Carwall software is said to prevent ‘mass hacks’ of vehicles’ on-board systems including those for connected and autonomous driving.  

Carwall sits in the vehicle ECUs and ‘learns’ the factory settings. If hackers breach the manufacturer’s cyber security and tries to infect the ECUs of in-service vehicles, Karamba’s software detects the impending change to factory settings and blocks activation.
 
David Barzilai, the company’s chairman and co-founder, said with tens of millions of lines of code in car software, it is impossible to guarantee all security bugs are eliminated. Carwall does not stop a hacker exploiting a security bug to transmit malware to a vehicle’s ECUs but it does prevent that malware being activated.

When Carwall detects foreign activity or code on an ECU it sends an alert to the manufacturer and system providers’ details on security bugs the hackers exploited, the code they attempted to run and the function it would execute. According to Barzilai, as the factory settings are definitive, Carwall does not produce false positives.

The software can be installed retrospectively to in-service vehicles by authorised distributers but cannot prevent individual hacks where the hacker can physically connect the vehicle’s CANbus architecture.

Related Content

  • April 27, 2015
    Rail signalling system ‘could be liable to hacking’
    A new rail signalling system to be installed across the UK could be liable to hacking, a government adviser has warned. Professor David Stupples told the BBC that the European Rail Traffic Management system (ERTMS) could be exposed to malicious software, or malware, and used to cause an accident perhaps telling the system the train is slowing when down when it is speeding up. "However, he said governments aren't complacent."Certain ministers know this is absolutely possible and they are worried about
  • September 8, 2015
    Over-the-air software updates to benefit for automotive market, IHS says
    While quite common in smartphones and personal computers, remote over-the-air (OTA) software updates are still only in their infancy in the automotive space, according to a new report from IHS Automotive. The report finds that OTA software updates will eventually be a big benefit for the automotive industry due to their capacity to reduce warranty costs, potentially increase overall completion rates for software-related recalls, improve customer satisfaction by eliminating trips to the dealership for so
  • May 21, 2013
    NHTSA studies hacking risks to automated vehicles
    A report by Bloomberg says that rising hacking risks to drivers as their cars become increasingly powered by and connected to computers have prompted the US’s auto-safety regulator to start a new office focusing on the threat. “These interconnected electronics systems are creating opportunities to improve vehicle safety and reliability, but are also creating new and different safety and cybersecurity risks,” David Strickland, head of the National Highway Traffic Safety Administration, said at a recent Senat
  • March 13, 2018
    ComNet introduces Port Guardian physical port lockout
    US communications networking equipment manufacturer ComNet has added the Port Guardian cyber-security features to its latest generation of self-managed switches which can physically disconnect a port if unauthorised access is detected. Port Guardian covers situations where network access is attempted by disconnecting an IP addressable device connected to the network. When Port Guardian senses intrusion, a notification is sent and the effected port is physically locked out, preventing access and thwarting