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

And in another parallel to conventional forms of attack, “states are probably one of the biggest risks here because they have the power and resources to invest in this. [type of attack],” said David Youssef, managing director at FTI Consulting and leader of the North American incident response efforts for the firm’s cybersecurity practice.

Actors’ bad motives for malware attacks are also common, according to security experts, who say hackers may target AI systems for the same reasons they launch other types of cyber attacks, such as causing disruption or harming an organization. Others say that hackers may use malware attacks to gain access to proprietary data or get money.

“Can someone use this to cheat? Absolutely,” said Erik Avakian, a technology consultant at Info-Tech Research Group and CISO for the former Commonwealth of Pennsylvania. “If a hacker can compromise the system by poisoning, they can use that; they would say, ‘We poisoned the model, now you have to pay us money [to get information on what we did].’”