Sunday, March 7, 2021
Home Tech EU report warns that AI makes autonomous vehicles ‘highly vulnerable’ to attack

EU report warns that AI makes autonomous vehicles ‘highly vulnerable’ to attack

The dream of autonomous vehicles is that they will keep away from human error and save lives, however a brand new European Union Agency for Cybersecurity (ENISA) report has discovered that autonomous vehicles are “highly vulnerable to a wide range of attacks” that might be harmful for passengers, pedestrians, and folks in different vehicles. Attacks thought of within the report embody sensor assaults with beams of sunshine, overwhelming object detection techniques, back-end malicious exercise, and adversarial machine studying assaults offered in coaching information or the bodily world.

“The attack might be used to make the AI ‘blind’ for pedestrians by manipulating for instance the image recognition component in order to misclassify pedestrians. This could lead to havoc on the streets, as autonomous cars may hit pedestrians on the road or crosswalks,” the report reads. “The absence of sufficient security knowledge and expertise among developers and system designers on AI cybersecurity is a major barrier that hampers the integration of security in the automotive sector.”

The vary of AI techniques and sensors wanted to energy autonomous vehicles will increase the attack floor space, in accordance to the report. To deal with vulnerabilities, its authors say policymakers and companies will want to develop a safety tradition throughout the automotive provide chain, together with for third-party suppliers. The report urges automobile producers to take steps to mitigate safety dangers by considering of the creation of machine studying techniques as a part of the automotive trade provide chain.

The report focuses on cybersecurity assaults with adversarial machine studying that carries the danger of malicious assaults undetectable to people. The report additionally finds that using machine studying in automobiles would require a steady assessment of techniques to guarantee they haven’t been altered in a malicious method.

“AI cybersecurity cannot just be an afterthought where security controls are implemented as add-ons and defense strategies are of reactive nature,” the paper reads. “This is especially true for AI systems that are usually designed by computer scientists and further implemented and integrated by engineers. AI systems should be designed, implemented, and deployed by teams where the automotive domain expert, the ML expert, and the cybersecurity expert collaborate.”

Scenarios offered within the report embody the potential for assaults on movement planning and decision-making algorithms and spoofing, like the sort that can idiot an autonomous car into “recognizing” automobiles, individuals, or partitions that don’t exist.

In the previous few years, a variety of research have proven that bodily perturbations can idiot autonomous car techniques with little effort. In 2017, researchers used spray paint or stickers on a cease signal to idiot an autonomous car into misidentifying the signal as a velocity restrict signal. In 2019, Tencent safety researchers used stickers to make Tesla’s Autopilot swerve into the incorrect lane. And researchers demonstrated final 12 months that they might lead an autonomous car system to shortly speed up from 35 mph to 85 mph by strategically inserting just a few items of tape on the highway.

 

The report was coauthored by the Joint Research Centre, a science and tech advisor to the European Commission. Weeks in the past, ENISA launched a separate report detailing cybersecurity challenges created by synthetic intelligence.

In different autonomous car news, final week Waymo started testing robo-taxis in San Francisco. But an MIT process pressure concluded final 12 months that autonomous vehicles might be at the least one other decade away.

VentureBeat

VentureBeat’s mission is to be a digital city sq. for technical decision-makers to acquire information about transformative expertise and transact.

Our web site delivers important info on information applied sciences and techniques to information you as you lead your organizations. We invite you to turn out to be a member of our neighborhood, to entry:

  • up-to-date info on the themes of curiosity to you
  • our newsletters
  • gated thought-leader content material and discounted entry to our prized occasions, resembling Transform
  • networking options, and extra

Become a member

Leave a Reply

All countries
117,262,992
Total confirmed cases
Updated on March 7, 2021 11:42 pm

Most Popular

Most Popular

Recent Comments

Chat on WhatsApp
1
Hello
Hello,
How can we help you?