gbhackers.com·10 september 2026

Hackers Can Turn Vulnerable LiteLLM AI Gateways Into Root Access and Cloud Credential Theft

Image for the article: Hackers Can Turn Vulnerable LiteLLM AI Gateways Into Root Access and Cloud Credential Theft
Image via gbhackers.com

Nearly 10% of internet-exposed LiteLLM AI gateways accepted the default master key sk-1234 or required no authentication, creating risks of LLMjacking, credential exposure, and root-level code execution. A scan of 3,074 instances found 294 systems vulnerable. This highlights the importance of securing AI gateways to prevent unauthorized access and data breaches. The findings were reported by GBHackers Security.