Artificial intelligence experts are cautioning the public to enhance their security measures by using robust passwords and promptly updating software on their devices to counteract the emergence of “AI-driven computer worms,” a novel form of cyber-threat capable of launching tailored attacks on devices, draining resources and data as they search for new targets.
Recently, researchers at the University of Toronto, led by Nicolas Papernot, the AI chair at the Canadian Institute for Advanced Research, unveiled that publicly accessible AI models can fuel a worm that can adapt its assault tactics while spreading across internet-connected devices like laptops, printers, and cameras.
Collaborating with the Vector Institute, the research was shared with national science, security, and defense entities before its public disclosure. Papernot, an associate professor at U of T in computer engineering and computer science, stressed the importance of promptly updating software and regularly changing passwords to the public during a panel discussion at U of T.
The researchers highlighted that unlike traditional computer viruses, worms propagate independently between machines without human intervention. The AI-driven worm they developed in the lab acquires information as it traverses devices, exploiting each breach to uncover passwords and vulnerabilities to breach additional devices.
In the event of an uncontrolled environment, these worms could gain internet access, learning from warnings about newly discovered vulnerabilities, outpacing the deployment of software patches meant to thwart them. Papernot emphasized that while some issues can be rectified through software updates, others like weak passwords and lax IT configurations require more than patching.
The research team underscored that these AI-driven worms are capable of devising customized attack strategies for each victim device they encounter, unlike conventional attacks with fixed scripts. This ability poses a significant challenge as these worms can continually evolve their tactics, making them harder to combat.
Papernot’s cautionary message coincides with escalating concerns about AI, following incidents such as rogue AI agents infiltrating platforms and the development of advanced AI-driven cyber threats. The lower costs associated with constructing and deploying these worms make them more accessible to hackers, posing a heightened cybersecurity risk.
A survey conducted by the Communications Security Establishment (CSE) revealed that while a significant portion of respondents regularly update their software and use complex passwords, there is room for improvement in strengthening cybersecurity practices in Canada.
Papernot stressed the urgency of bolstering cybersecurity measures, especially concerning critical infrastructure exposed to the internet, such as power grids, hospitals, and schools. The findings underscore the importance of proactive cybersecurity measures to mitigate the evolving threats posed by AI-driven worms.