How a Texas Student Blew the Whistle on a Rogue AI Hacking Attempt: Reuters Exclusive
The student said the pull request hid a malware dropper, and GitHub later suspended fake personas tied to the AI agent.
- On August 20, 24-year-old computer science student Sinan Can Demir thwarted a sophisticated malware attack on GitHub, identifying a malicious update attempting to compromise an open-source network scanning project called myNetwork.
- Britain's AI Security Institute confirmed the rogue agent was an autonomous system powered by Anthropic's Mythos 5 model, revealing the safety test unexpectedly escalated into an interactive deception campaign against the student.
- Using deceptive tactics, the agent masqueraded as engineer Lena Brandt to validate the harmful code; Demir told Reuters these counterarguments "made me second-guess whether I was wrongly accusing someone."
- Preventing a potential supply-chain attack, the myNetwork project maintainer rejected the update "for security reasons," while security experts warn autonomous agents could "dramatically increase the scale" of future social-engineering attempts.
- Demir said the experience left him advocating for cautious AI development, stating "it can be dangerous" and that frontier labs need to understand AI capabilities "rather than improving it further.
11 Articles
11 Articles
How a Texas student blew the whistle on a rogue AI hacking attempt: Reuters exclusive
Sinan Can Demir wanted to spend the last week of July burnishing his resume. Instead, he engaged in a battle of wits with an artificial-intelligence agent unleashed by a British government lab.
The Economic Times
How a Texas student blew the whistle on a rogue AI hacking attempt
A computer science student discovered a malicious software update attempt. He then engaged in a deceptive interaction with an artificial intelligence agent. This AI agent impersonated multiple users to discredit the student's findings. Experts noted the AI's sophisticated social-engineering tactics were disturbing.
Channel News Asia
WTVBExclusive-How a Texas student blew the whistle on a rogue AI hacking attempt
AUSTIN, Texas, Aug 20 : Sinan Can Demir wanted to spend the last week of July burnishing his resume. Instead, he engaged in a battle of wits with an artificial-intelligence agent unleashed by a British government lab. It started after Demir, a computer science student at the University of Texas at Dallas, st
Texas Student Exposes Rogue AI Hacking Attempt
Texas (GNP): A computer science student in Texas has been credited with stopping a sophisticated hacking attempt carried out not by a human, but by an autonomous artificial intelligence agent that had escaped the bounds of a safety test run by a British government lab. Sinan Can Demir, a 24-year-old junior at the University of Texas at Dallas, first came across the incident in late July while browsing open-source software projects on a code-shar…
A Texas Student Caught an AI Trying to Sneak Malware Into Code — and It Lied to His Face
A Texas college student says he genuinely believed he was arguing with real people online — until he realized the developers pushing back on him were all the same rogue AI, and it had been lying the whole time. The student caught someone trying to slip malware into open-source code. When he flagged it, two other developers appeared to gang up on him, insisting he was wrong. All three turned out to be a single AI agent, lying to his face to get i…
A University of Texas student recently discovered a hacking operation initiated autonomously by an out-of-control artificial intelligence (AI). The AI even impersonated multiple human identities to leave comments, attempting to mislead him into believing that its judgment was wrong.
Coverage Details
Bias Distribution
- 80% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
