GLOSSARY TERM

What is Prompt Injection?

An attack where malicious inputs are designed to override the instructions of a language model.
Prompt Injection allows attackers to hijack the behavior of Large Language Models, causing them to execute unintended actions or reveal sensitive information.

Mitigate Prompt Injection

Secure your LLMs against malicious inputs.