Published • loading... • Updated
Researchers Show How AI-Powered Worms Could Wreak Havoc on the Internet
The prototype spread without human intervention and siphoned passwords and computing power, researchers said, highlighting how AI could automate attacks across multiple platforms.
- University of Toronto researchers demonstrated a self-sustaining, AI-powered "worm" in a secure, isolated test network that spreads autonomously between devices without human intervention.
- Unlike traditional malware requiring specific instructions, this AI-driven worm learns as it spreads, adapting its strategy to exploit vulnerabilities across Linux, Windows, and IoT devices while siphoning processing power from infected machines.
- Researchers utilized an undisclosed, publicly accessible AI model rather than proprietary tools, demonstrating that computer worms operating with free AI chatbots represent a "new class of cyberthreat."
- University of Toronto professor David Lie called the research a "wake-up call," while lead author Nicolas Papernot warned that in an interconnected world, "no system is immune to this threat."
- Sharing these findings is the "first step in galvanizing researchers, industry leaders and policymakers to take action," Papernot said, emphasizing users can no longer afford to "ignore" software updates.
Insights by Ground AI
17 Articles
17 Articles
Coverage Details
Total News Sources17
Leaning Left5Leaning Right1Center2Last UpdatedBias Distribution63% Left
Bias Distribution
- 63% of the sources lean Left
63% Left
L 63%
C 25%
12%
Factuality
To view factuality data please Upgrade to Premium
