This research explores how generative AI can dramatically lower the barrier to entry for robot hacking.
Using Cybersecurity AI (CAI) we analyzed three consumer robots:
– robotic lawn mower
– powered exoskeleton
– window-cleaning robot
In about 7 hours the system identified 38 vulnerabilities including firmware exploitation paths, BLE command injection and unauthenticated root access.
Historically this type of analysis required weeks of specialized robotics security research.
Would be really interested to understand the cost to complete the reviews, and the man time cost in getting to the actual vulns and getting rid of all the false positives. did you come across many false positives?
Hi HN – one of the authors here.
This research explores how generative AI can dramatically lower the barrier to entry for robot hacking.
Using Cybersecurity AI (CAI) we analyzed three consumer robots: – robotic lawn mower – powered exoskeleton – window-cleaning robot
In about 7 hours the system identified 38 vulnerabilities including firmware exploitation paths, BLE command injection and unauthenticated root access.
Historically this type of analysis required weeks of specialized robotics security research.
Happy to answer questions.
Would be really interested to understand the cost to complete the reviews, and the man time cost in getting to the actual vulns and getting rid of all the false positives. did you come across many false positives?
[dead]