U of T researchers discover AI worm that can spread across online devices and hijack networks
By
Adina Bresge
FeedBagel synthesis
· 2 sourcesResearchers from the University of Toronto, Vector Institute, University of Cambridge, and ServiceNow have developed a proof-of-concept AI-powered worm that can autonomously adapt and spread across computer networks. Hacker News reported that the worm, built using free, publicly accessible AI models, can seize control of entire networks and hijack computing power. According to coverage on bsky, in seven-day autonomous lab runs, the worm averaged 31.3 vulnerabilities found and 23.1 hosts exploited, with the authors warning that self-sustaining AI-driven cyber threats are no longer theoretical.
An everything bagel for the brain. Substantive, layered, well-seasoned.
Summary
University of Toronto researchers have discovered a new class of cyberthreat — an AI-powered worm that can adapt its strategy as it spreads between devices. Built using free, publicly accessible AI models, this worm can seize control of entire networks and hijack computing power, posing a significant challenge to current cybersecurity defenses. The research, released June 2, is believed to be the first demonstration of such a threat.
Key quotes
· 3 pulledThe researchers, who released their work June 2, are believed to be the first to show that publicly accessible AI models can be used to power a worm that adapts its strategy as it spreads from one device to the next.
It can seize control of an entire network and hijack computing power to allow hackers to launch sophisticated attacks.
Every online device is a potential target. And current cyber defences are not yet ready for it.
You might also wanna read
Researchers demonstrate AI-powered malware worm that adapts attacks across computer hosts
Researchers from the University of Toronto, Vector Institute, University of Cambridge, and ServiceNow have developed a proof-of-concept AI-d
winbuzzer.com·5h ago
How generative AI-powered hacking tools are reshaping the cyber attack landscape and defense strategies
The article examines how generative AI has transformed the cyber threat landscape since WormGPT's emergence in June 2023. AI-powered hacking
hendryadrian.com·6d ago
TuxBot V3: LLM-Powered IoT Malware from AISURU/Keksec Group Discovered by Unit 42
Palo Alto Networks Unit 42 has discovered TuxBot v3, an IoT malware framework associated with the AISURU/Keksec threat group. This new varia
undercodetesting.com·2d ago
AI bug-finding systems uncover real vulnerabilities at DARPA cybersecurity challenge
The article discusses the DARPA AI Cyber Challenge (AIxCC) held in Las Vegas, where top cybersecurity teams demonstrated AI-powered bug-find
The Verge·1mo ago
Predictive Multicast Tech Wins Intel Award with 1.86× AI Speedup, Raises Network Security Concerns
Professor Jiayi Huang of HKUST(GZ) received Intel's 2025 Outstanding Researcher Award for developing a predictive multicast technology that
undercodetesting.com·13h ago
Global Powers Race for Humanoid Robot Dominance Amid Rising Cyber Threats
The article discusses how global superpowers are racing to develop embodied AI systems (humanoid robots) while simultaneously using cyberatt