RESEARCHERS at the University of Toronto have developed a new category of autonomous malware, termed "AI Worms," capable of adapting its attack strategies based on the specific devices it targets, including Windows, Linux, and IoT devices. The malware uses open-source AI models to observe and dynamically generate tailored compromise strategies, making it more efficient than traditional worms, which rely on specific vulnerabilities.
This innovative capability allows it to evade typical cybersecurity measures by exploiting a range of weaknesses in networks, deploying resources from infected machines to facilitate further attacks. The study emphasizes the need for evolving defensive strategies against such adaptable threats, highlighting the importance of maintaining robust cybersecurity practices.