Artificial intelligence experts are cautioning the public to enhance their online security measures by using robust passwords and promptly updating software on their devices to counteract the threat of “AI-driven computer worms.” These advanced cyber-threats can launch tailored attacks on devices, depleting their resources and stealing data as they seek out new targets.

A recent study by the University of Toronto, led by Nicolas Papernot, the AI chair at the Canadian Institute for Advanced Research, revealed that publicly available AI models can power adaptable worms capable of adjusting their attack strategies in real-time as they move across internet-connected devices like laptops, printers, and cameras.

Collaborating with the Vector Institute, the research findings were shared with key national science, security, and defense agencies before publication. Papernot, an associate professor at U of T specializing in computer engineering and computer science, emphasized the importance of promptly updating software and regularly changing passwords to safeguard against potential threats.

The AI-driven worms, unlike traditional viruses, autonomously spread between machines without human intervention, gathering intelligence as they traverse devices. They exploit vulnerabilities and weak points to infiltrate new systems, making them difficult to contain with conventional security measures.

Papernot highlighted the critical need for improved cybersecurity practices, stressing the importance of multi-factor authentication, timely software updates, and organizational adaptation to ensure swift deployment of security patches. These worms are not only more effective but also cost-efficient to develop and deploy, posing a significant challenge to traditional defense mechanisms.

The study underscores the urgency for enhanced cybersecurity measures in Canada, particularly in critical infrastructure sectors such as power grids, healthcare facilities, and other essential services vulnerable to online threats. Papernot’s research sheds light on the pressing need for proactive cybersecurity efforts to mitigate the risks posed by evolving AI-driven cyber threats.