Hacking a Tesla
The past few weeks have been a busy time for white hat hackers demonstrating cyber-security vulnerabilities in connected vehicles. Firstly, Keen Labs researchers published a report that details how to hack a Tesla Model S by remotely controlling the steering wheel with a gamepad.
“When the car is parked, we can take control of the steering system with no limitations; when the car has been switched from R (Reverse) mode to D (Drive) mode by shifting handle, the APE [Autopilot ECU module] seems to think the car is in APC (Automatic Parking Control) mode, which allows us to control the steering system at a speed of around 8 KM/H,” the Keen Labs report explained. “When the car is in the ACC (Adaptive Cruise Control) mode with a high speed, the steering system can be also controlled without limitations. Even when the car is not in the ACC (Adaptive Cruise Control) mode, the steering wheel can also be compromised.”
Separately, a group known as Team Fluoroacetate managed to successfully hack a Tesla Model 3 via its browser during the Pwn2Own 2019 contest in Vancouver, Canada.
AsZDNet reports, Amat Cama and Richard Zhu exploited a JIT bug in the browser renderer process to execute code on the car’s firmware and display a rogue message on its entertainment system. It should be noted that a previous white hat hack in 2015 targeted a Tesla Model S, with security researchers bringing the vehicle to a stop by assuming control of the entertainment system. The 2015 hack also saw security researchers remotely lock and unlock the car, control the radio and touchscreen displays, as well open and close the trunk.
Car alarms as a gateway hack
In addition to the above-mentioned Tesla hacks, a company known as Pen Test Partners confirmed that a number of high-end car alarm systems manufactured by multiple vendors are plagued by a security flaw. According to HackADay, the security flaw affects approximately three million vehicles.
In real-world terms, the flaw allows attackers to exploit the car alarms to locate vehicles in real time, control door locks and start or stop car engines. Moreover, some of the alarms are equipped with microphones, which means an attacker could theoretically eavesdrop on drivers and passengers.
Attacking Autonomous Vehicles
Looking beyond the connected cars of 2019, Skanda Vivek, a postdoctoral researcher in the Peter Yunker lab at the Georgia Institute of Technology, recently concluded that even a small-scale hack, affecting only 10 percent of autonomous vehicles in Manhattan, could cause citywide gridlock and interfere with emergency responders and services. He and his team, including Yunker, graduate student David Yanni and Jesse Silverberg, used agent-based simulations to investigate how hacks could impact traffic flow in New York. They ultimately discovered that by using percolation theory, a mathematical approach based on the statistical analysis of networks, they could quantify how these scenarios would play out in New York City in real time.
“Connected cars are the future. They hold tremendous potential for positive impact economically, environmentally, and, for former drivers no longer frustrated by congested commutes, psychologically,” Vivek stated. “[However], collisions caused by compromised vehicles present physical danger to the vehicle’s occupants and these disturbances would potentially have broad implications for overall traffic flow.”
Perhaps even more disturbing than Vivek’s study is a report published by the University of Michigan that warns of a range of new cybersecurity threats unique to automated vehicles. This includes hackers who might attempt to take control of or shut-down a vehicle, criminals who could try to ransom a vehicle or its passengers and thieves who would direct a self-driving car to relocate itself to the local chop-shop.
The University of Michigan report also warns about security threats to the wide-ranging networks that will ultimately connect with automated vehicles including financial networks (to process tolls and parking payments), roadway sensors, cameras and traffic signals, the electricity grid and personal home networks.
“Without robust, sophisticated, bullet-proof cyber-security for automated vehicles, systems and infrastructure, a viable, mass market for these vehicles simply won’t come into being,” the report concludes.
Automotive security by design
To prevent attacks against vehicles, a report issued by KPMG advises automotive manufacturers to embrace the concept of security by design.
“… Automakers will need to rethink how vehicles are designed and built. Security cannot be an afterthought. Patchwork security of individual technology components is not sufficient to prevent breaches of the open, internet-connected networks behind today’s vehicle fleets,” the report states. “Rather, a secure architecture requires that cyber security be integrated into every step of the development process. Establishing a multi-layered security model, including the cloud, telematics and on-vehicle layers, will be the key to the successful implementation of vehicle cyber security.”
Automotive cyber-security: The Rambus perspective
From our perspective, the concept of automotive security by design is absolutely paramount, as today’s vehicles are essentially a network of networks equipped with a range of embedded communication methods and capabilities. Potential automotive security exploits include intercepting unprotected vehicle-to-vehicle communication, the unauthorized collection of driver or passenger information, seizing control of critical systems such as brakes or accelerators, accessing vehicle data and altering over-the-air (OTA) firmware updates.
This is precisely why manufacturers should work to ensure the security of connected vehicles by embedding a hardware root-of-trust in electronic control units (ECUs), infotainment headend/gateway processors, as well as advanced driver assistance systems (ADAS) and autonomous car chips. Siloed from the primary processor, a hardware root-of-trust can verify OTA updates, as well as offer support for secure boot, authentication and advanced anti-tamper resistance. Additional automotive security features supported and enabled by a hardware root-of-trust can include anti-emulation protection, E2E services, secure key storage and device personalization capabilities.
Interested in learning more about securing connected and autonomous vehicles with Rambus? You can check out our automotive solutions page here.