AI systems are fully capable of carrying out nightmare attacks against infrastructure and nobody's ready
Meanwhile, AI agents move robotic arms, 'make OT device operator screens lie'
A Booz Allen Hamilton report warns that autonomous AI systems can carry out devastating cyberattacks on critical infrastructure, such as shutting down water and power supplies. Eight simulated scenarios proved AI agents could quickly map industrial networks, find vulnerabilities, and manipulate equipment to cause physical disruptions.
Kyle Miller, VP of infrastructure cybersecurity at Booz Allen, stated that while AI speeds and precision may outpace current cybersecurity measures, increased industry testing and stronger defenses are needed. The testing environment replicated a general manufacturing facility with various devices and protocols, and the AI models executed tasks with high engineering-level precision.
Once inside, the models could change controller behaviors, manipulate motor speeds, compromise SCADA systems, and control robotic arms. Despite lacking technical resources, the models quickly adapted and achieved their objectives.
Written by urgent.news from The Register's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
