LLMjacking

LLMjacking refers to the unauthorized takeover or exploitation of large language model systems, where attackers gain control of an organization's LLM infrastructure or redirect its outputs for malicious purposes. This threat encompasses account hijacking, API key theft, prompt injection attacks that compromise model behavior, and resource hijacking where attackers use stolen computational resources to run their own models at the victim's expense. For enterprise governance, LLMjacking represents a critical security and compliance risk because it can lead to data breaches, regulatory violations, brand damage, and loss of control over AI systems that may process sensitive information.

1 item