McAfee Research Highlights Tesla Driverless Tech Vulnerabilities

A Tesla Model S and Model X are displayed at a shopping mall in Hong Kong

A Tesla Model S and Model X are displayed at a shopping mall in Hong Kong

They found that adding even a 2-inch horizontal piece of black electrical tape to the middle bar of the "3" in 35 led to a misclassification of 85 miles per hour.

They ultimately determined that with Tesla's Speed Assist and Tesla Automatic Cruise Control features enabled, the modifications to the signs could result in the vehicle accelerating and exceeding the speed limit on its own.

After sticking the tape, the researchers rode the Tesla on the highway.

The flaw is said to have caused 110 crashes and 52 injuries, with many drivers stating the incident occurred when they attempted to park in a garage or at a curb.

As we briefly mention above, McAfee Advanced Threat Research has been studying what we call "Model Hacking", also known in the industry as adversarial machine learning.

Tesla vehicles are equipped with Mobileye EyeQ3 camera systems which read the speed limit signs and help with setting the accurate speed.

That said, it is worth noting that this is not the first attack on Tesla " s AI system. Ultimately, we were successful in creating extremely efficient digital attacks which could cause misclassifications of a highly robust classifier, built to determine with high precision and accuracy what it is looking at, approaching 100% confidence.

Speaking to MIT Tech Review, McAfee researcher Povolny stated that the results of the research are worrying, and that many 2016 Tesla vehicles are on the road.

'Parasite' distributor claps back at Donald Trump
Pitt mentioned John Bolton , joking about how the Senate voted against allowing witnesses during Trump's impeachment trial . It happened again Thursday night in Colorado Springs, where his speech somehow meandered to the Oscars.

Yeah, so a group of McAfee researchers did just that - manipulated a speed limit sign that tricks the Autopilot self-driving technology that Tesla is a leader in.

And MobilEye EyeQ3 dismissed the research altogether.

Speaking to MIT Tech Review, a representative said that the changed signage can be read wrong by people and that the camera is not designed for full autonomous driving. While semi-autonomous vehicles have existed for many years, the vision of fleets of fully autonomous vehicles operating as a single connected entity is very much still a thing of the future. The tape made the "3" look like an "8".

"Simply by perturbating - changing the magnitudes of a few features (such as pixels for images), zeros to ones/ones to zeros, or removing a few features - the attacker can wreak havoc in security operations with disastrous effects", McAfee researchers Steve Povolny and Celeste Fralick wrote in a blog post describing the attack. The researchers managed to deceive Tesla's speed limit.

The tests were carried out using a 2016 Tesla Model X and Model S, which used camera systems from Mobileye. Of course, they knew the vehicle was going to do this.

The Tesla stopped when it hit the garage's concrete wall.

The owner of the Model X, Puzant Ozbag, said the vehicle was only five days old and his wife, who was behind the wheel at the time of the incident, had not activated any self-driving features at the time of the crash.

Recommended News

We are pleased to provide this opportunity to share information, experiences and observations about what's in the news.
Some of the comments may be reprinted elsewhere in the site or in the newspaper.
Thank you for taking the time to offer your thoughts.