AI system poisoning is a growing threat — is your security regime ready?

And in one other similarity to traditional assault varieties, “nation-states are in all probability one of many largest dangers right here as a result of they’ve the flexibility and sources to speculate on this [type of attack],” says David Youssef a managing director at FTI Consulting and chief of the North America incident response efforts for the agency’s cybersecurity apply.

Dangerous actor motivations for poisoning assaults are additionally acquainted, in line with safety specialists, who say hackers might goal AI methods for a similar causes they launch different kinds of cyberattacks, comparable to to trigger disruption or injury a company. Some say hackers may additionally use poisoning assaults to realize entry to proprietary knowledge or to get cash.

“May somebody use this for extortion? Completely,” says Erik Avakian, technical counselor at Information-Tech Analysis Group and former state CISO for the Commonwealth of Pennsylvania. “If a hacker can compromise a system by poisoning, they will use that; they will say, ‘We poisoned the mannequin, now you must pay us cash [to get information on what we did].’”