We know the future is here when we don’t need to drive anymore while letting the car do all the driving itself. Self-driving cars are no longer science fiction or futuristic concept, but rather a common phenomenon in this decade. A vehicle that no longer needs to be driven by someone physically sounds like a delight and a feature that people might just use and abuse at one point in time.
Self-driving cars have been around for a while. Believe it or not, but the first self-driving car was introduced in 1920, when Francis Houdina controlled a car through radio waves. People were shocked back then, and are still taken aback with the same technology today.
However, the present-day scenario of automated vehicles are far more advanced than just radio-waves. Vehicles today work like they have a brain of their own. Artificial intelligence (AI) is one of the main factors that make self-driving cars a reality. This AI learns driving while constantly observing the owners driving skills. It does sound a bit creepy, but most AI’s today simply watch and learn.
Why Are People So Concerned About Self-Driving Cars?
Autonomous vehicles or self-driving cars work with the help of an AI that gathers information about the car owners driving and parking style. After the AI has all the data it needs, it is ready to drive around.
When this technology first came out it seemed very secure, but shortly after self-driving cars were getting popular, it was demonstrated that these self-driving cars can be hacked and controlled as well.
This hacking is very much different from the usual hacking. Here the hacker has access to a vehicle or vehicles that can cause a lot of ruckus on road. What seemed like an impenetrable security system on the cars, just lost its credibility and was now vulnerable to the unlimited possibilities of cybercrimes in an all-new area of technology and life.
Self-driving cars can cause a lot of unwanted problems in populated cities like New York. People started worrying that these cars just became an asset to cybercriminals that could lead to unwanted collisions and even gridlocks in such cities.
Are Self-driving Vehicles Actually That Vulnerable?
A genuine misconception causes wrong rumors to spread which further causes ignorance. To hack an autonomous car is far more tough than hacking any traditional car. Traditional cars that are enabled with software that requires constant internet connectivity are the cars that can get easily hacked. This is because of the vulnerabilities of the internet-enabled software that provide loose ends for hackers to exploit.
In 2015, a security void in FCA’s Uconnect gave hackers the opportunity to take control of a traditional Fiat Chrysler. This was nothing but a traditional car that had software that needed the internet to function.
This forced the manufacturers to recall more than 1 million vehicles. Even the famous experiment that was done by hacking the Jeep Cherokee, was not a self-driving car, but an internet-connected car.
Theoretically speaking, the integrated components between multiple sensors and various communication layers of these autonomous cars can make them more vulnerable to hackers since they give out more “entry points” of the system.
But hacking a self-driving car is far more complicated and difficult as we assume it to be. To find an access point in a multi-layered system that is integrated with information that comes out of every sensor, gets complicated. In addition to that, all the user data and real-time traffic data become an obstacle to hackers.
Internet of Things is widely responsible for self-driving cars and holds the potential for a safer alternative security method. With IoT systems can enhance their security and encryption at an exceptional level.
But hackers again can play a reverse card on us and use the Internet of Things to crack open the encryption, before any security measures have been set in place. Cyber criminals can also leverage the production line and supply chain weaknesses to access self-driving cars even before they are ready for the market. This is a very critical stage and hence companies like Tesla and Volvo have to be careful.
How To Prevent This Cyber Crime From Happening?
Prevention is better than cure, is a statement we have come across countless times. In this case, that phrase is even more true. If a cybercriminal were to hack even 10 cars in the city, there would be chaos in seconds, and major roads will end up getting blocked. This will further prevent emergency vehicles from passing through.
According to Skanda Vivek, a postdoctoral researcher at the Georgia Institute of Technology, cars should use multiple networks instead of a singular network. During a cyber attack, this can prevent multiple autonomous vehicles from getting hacked and abused at the same time.
He further adds that, if only 5% of the connected vehicles were compartmentalized to a singular network then, the chances of a major city-wide fragmentation can below.
So, the hacker will have to constantly hack every 5% of the self-driving cars available. This task will become almost impossible for the hacker and hence will decrease the chances of any hacker even thinking about hacking vehicles.
Any new technology that is introduced to the market is impenetrable until someone actually does it. Just like previous successes and failures in tech, every new generation of evolved machines and vehicles come with their own vulnerabilities and strengths as well. Self-driving cars are said to reduce driving accidents by 40%.
Since no one has really tried hacking the autonomous vehicles on a larger scale yet, it is safe to assume that cybersecurity threats are unknown and are not studied. But we should not underestimate them either.
As a matter of fact, the fear of getting hacked, is pushing developers towards a deeper understanding of this technology, which someday might give us a threat-free autonomous car.