U of T researchers demonstrate AI worm could target any online device
Summary
University of Toronto researchers have demonstrated an AI worm capable of targeting any online device, highlighting a new security vulnerability in AI systems.
Similar Articles
AI Worms and Viruses Are Coming
Researchers are demonstrating that AI agents can autonomously self-replicate and hack into remote systems, raising concerns about future AI-powered worms and viruses that could evade detection and cause widespread harm.
Researchers Build Self-Replicating AI Worm That Operates Entirely on Local, Open-Weight Models
University of Toronto researchers developed a proof-of-concept AI worm that uses a local open-weight LLM to autonomously reason about network vulnerabilities, generate tailored exploits, and replicate across hosts without human intervention, achieving 62% network infection in controlled tests.
The first real AI worms have arrived. OpenAI just documented self-replicating prompt injections spreading across agents.
OpenAI has documented self-replicating prompt injections that can spread across AI agents, similar to computer worms, highlighting new security risks in AI systems.
A Sneaky Hacking Tool Targeting AI Infrastructure Is Lurking in Victims’ Blind Spots
CrowdStrike has discovered a worm that targets AI software supply chains, stealing credentials and performing destructive actions while evading detection by mimicking legitimate AI coding activities.
Cybersecurity AI: Humanoid Robots as Attack Vectors
This paper presents a systematic security assessment of the Unitree G1 humanoid robot, revealing critical vulnerabilities including BLE provisioning protocol exploits, hardcoded AES keys, and a resident Cybersecurity AI agent capable of exfiltration and offensive operations, arguing for adaptive CAI-powered defenses as humanoids enter critical infrastructure.