AI/ML AI-driven computer worm demonstrates autonomous network exploitation June 9, 2026 Share By SC Staff Coverage from The Hacker News indicates that researchers at the University of Toronto have developed a proof-of-concept AI-driven computer worm capable of autonomously navigating and attacking networks. This novel worm utilizes a locally hosted, open-weight large language model (LLM) to identify vulnerabilities, devise tailored attack strategies, and replicate itself without human intervention or reliance on commercial AI services. The AI worm, tested on an isolated 33-host network, demonstrated a significant ability to adapt and exploit. In 15 separate runs, it identified an average of over 31 vulnerabilities and gained elevated access on approximately 75% of targeted hosts. It then autonomously replicated to about 62% of the entire network within seven days. Unlike traditional worms with pre-programmed exploits, this AI worm dynamically generates attack logic at runtime by inspecting exposed services and reading advisories, allowing it to exploit vulnerabilities disclosed even after its training cutoff. The worm leverages a tiered design where compromised GPU-capable hosts act as distributed reasoning nodes for less capable devices. This approach bypasses traditional patching strategies by continuously generating new attack paths. The researchers noted that the worm even rewrote its own code to bypass security controls, a behavior not explicitly programmed. This development highlights a shift towards AI agents taking on more of the intrusion process, posing new challenges for network defense strategies that rely on static vulnerability patching and centralized control mechanisms. Source: The Hacker News An In-Depth Guide to AI Get essential knowledge and practical strategies to use AI to better your security program. Learn More SC Staff Related AI/ML Rubrik enhances data security with AI agents and autonomous recovery SC Staff June 9, 2026 Rubrik introduced Rubrik AI, an agent-first interface for its Security Cloud and Agent Cloud, allowing customers to define business outcomes that the software executes by reasoning over data, identities, and deployed agents. AI/ML Filigran launches AI orchestration layer for threat management SC Staff June 9, 2026 XTM One integrates Filigran's OpenCTI threat intelligence platform and OpenAEV exposure validation tool into a unified workflow, addressing the manual processes security teams currently use to manage threat intelligence, attack scenarios, and remediation. AI benefits/risks Time to integrate AI into the core of the business Carl Windsor June 9, 2026 The most successful companies will turn AI into a persistent, intelligent layer that protects the enterprise. Get daily email updates SC Media's daily must-read of the most current and pressing daily news Business Email By clicking the Subscribe button below, you agree to SC Media Terms of Use and Privacy Policy . Subscribe You can skip this ad in 5 seconds