Hacker Shows How to Turn a Tesla Model S Into a ‘Surveillance Station’
A security researcher created a solution that utilizes ALPR software and a machine learning network to spot, track and store license plates and faces.
Tesla vehicles are known for a lot of things, for better or for worse. They are marketed as being good for the environment and coming loaded with advanced technology.
However, features like its autopilot function can lead to drivers paying less (or no) attention to the road. Being covered in numerous cameras and sensors also opens the door to hacking, in both good ways and bad.
We’ve seen a security dealer use his Tesla to automate his home. Now, we have a security researcher that has turned his Tesla Model S into what he calls an “AI-powered surveillance station.”
Several weeks ago at the DEF CON hacker conference in Las Vegas, Truman Kain unveiled the Surveillance Detection Scout, a DIY computer that fits into the console of a Tesla Model S or Model 3 and plugs into the dashboard USB port.
The computer turns the car’s built-in cameras into a surveillance system that spots, tracks and stores license plates and faces. The system can then push an alert to the car touchscreen and user’s phone if it repeatedly sees the same license plate.
When parked, it will track nearby faces to see which repeatedly appear. The intent is to offer a warning that someone might be preparing to steal the car, tamper with it, or break into the driver’s nearby home, says Kain.
Kain was able to create the Surveillance Detection Scout primarily by using free code available on GitHub (Kain has made Surveillance Detection Scout available on GitHub as well).
The program works by utilizing Tesla’s three cameras — two of which are located in the sideview mirrors and one forward-facing — on a Nvidia Jetson Xavier mini-computer.
Its machine learning engine is an open source neural network framework called Darknet, which is paired with ALPR Unconstrained for recognizing license plates and Facenet for tracking faces. The system also uses Google’s Open Images Dataset as training data.
“I’m not doing any cutting-edge AI,” Kain says. “I’m just applying what’s already freely available, off the shelf.”
Of course, being able to hack a Tesla so easily for this type of use raises ethical and privacy concerns. Could we see a sudden rise of nefarious surveillance Teslas cruising around town? Technology and ideas that were once science fiction are now becoming part of our everyday lives.
You can read more about the Surveillance Detection Scout and the concerns it raises, here. Check out a demo below.
Security Is Our Business, Too
For professionals who recommend, buy and install all types of electronic security equipment, a free subscription to Security Sales & Integration is like having a consultant on call. You’ll find an ideal balance of technology and business coverage, with installation tips and techniques for products and updates on how to add sales to your bottom line.
A free subscription to the #1 resource for the residential and commercial security industry will prove to be invaluable. Subscribe today!