AI Dynamics

Global AI News Aggregator

About

Differentiating AI Jailbreaks from Prompt Injection Attacks

What you’re describing is a jailbreak, not a prompt injection. Prompt injection is when data is misinterpreted as part of the prompt instructions (against the intention of the prompter). A jailbreak is when the prompter bypasses safety policies of the model.

→ View original post on X — @goodside