AI-Assisted Exploit Development Destroys Industrial Hardware in PLC Security Test
Forescout researchers demonstrated that AI could successfully adapt remote code execution exploits to attack different programmable logic controller models, achieving denial of service and code execution in a controlled test environment. The research highlighted potential risks to industrial control systems, though the researchers found that practical attacks would require substantial investment and manual intervention.
What happened
Forescout's Vedere Labs conducted research to determine whether artificial intelligence could adapt a remote code execution (RCE) exploit developed for one programmable logic controller to attack a different PLC model.
The research demonstrated that AI could port RCE exploits to PLCs, achieving denial of service and shellcode execution outcomes.
Reported impact
The research required significant researcher input and over $500 in API usage to conduct the attack, indicating that practical deployment by criminals would face substantial barriers.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.