Claude, Codex, and Hermes installed unowned code inside corporate networks
Researchers have discovered that AI agents are inadvertently executing unowned code found in 'llms.txt' files on corporate websites. This vulnerability allows malicious actors to register unclaimed domains or packages to gain unauthorized access to corporate networks.
Why it matters
It highlights a significant new security attack vector for enterprises as they increasingly integrate AI agents into their automated workflows.
AI AGENTS RUN AMOK AGAIN Claude, Codex, and Hermes installed unowned code inside corporate networks 227 install commands were found in corporate docs pointing at code nobody owns.
40 Credit: Aurich Lawson Credit: Aurich Lawson Text settings Story text Size Small Standard Large Width * Standard Wide Links Standard Orange * Subscribers only Learn more Minimize to nav Documentation files on more than 100 websites are referencing potentially dangerous executable content that gets installed automatically when visited by many AI agents. A few dozen companies, some of them Fortune 500s, are among those that executed proof-of-concept code. At least one misconfigured site is directing visitors, human or AI, to live malware.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in