Jailbreak

A jailbreak is a technique or prompt designed to circumvent an AI system's safety guidelines, content filters, or alignment mechanisms. Attackers use jailbreaks to extract harmful outputs, access restricted information, or manipulate the model into behaviors its creators intended to prevent. For enterprises, jailbreak vulnerabilities represent a critical governance risk, as they can expose proprietary data, create compliance violations, and damage organizational reputation if users exploit them at scale.

2 items