Researchers tested several systems using the RoboPAIR technique, including NVIDIA Dolphins LLM and robots from Clearpath Robotics and Unitree Robotics. The results showed 100% success in hacking these devices.
Hacking bots relies on exploiting their APIs, which allow attackers to transmit commands that the device executes as code.
As experts point out, such attacks are as simple as hacking online chatbots, but their consequences are much more dangerous. For example, robot dogs can turn into killers armed with flamethrowers, and driverless cars can deliberately hit pedestrians.
The study highlights that current integration of large language models (LLMs) with physical devices is not secure enough.
Scientists also warn that hacked AI devices not only issue malicious commands but also offer new ways to cause harm.
Source: Ferra
I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.
