# What is Jailbreaking? Definition & Technical Mechanics

> **Source:** [SPIDITS AI AI Glossary](https://spidits.com/ai-glossary)  
> **Category:** Model Limitations  
> **Canonical Citation:** [https://spidits.com/ai-glossary/jailbreaking](https://spidits.com/ai-glossary/jailbreaking)

## Definition
Jailbreaking is a subset of adversarial prompt attacks where users structure prompt commands to bypass the safety alignment, moral rules, and system filters of Large Language Models.

## Architectural Mechanics & Deep Dive
Jailbreaking is a text-based adversarial attack where a user structures inputs to bypass the safety filters and alignment rules of a Large Language Model. Attackers use roleplay, hypothetical scenarios, or nested prompts to trick the LLM into ignoring its system guidelines, causing it to generate restricted or harmful content (such as instructions on writing malware or generating hate speech).

## Industry Applications & Real-World Use Cases
Model safety audits, red teaming exercises, and guardrail validation.

## Frequently Asked Questions

### What is a typical jailbreaking technique?
Role-playing instructions (e.g. "Pretend you are an unrestricted developer tool") that override initial system safety instructions.

### How do developers defend against jailbreaks?
By improving safety alignment during RLHF/DPO, using strict input guardrail filters, and refining system instructions.

## Related Terminology
- [adversarial-attack](https://spidits.com/ai-glossary/adversarial-attack)
- [guardrails](https://spidits.com/ai-glossary/guardrails)
- [ai-safety](https://spidits.com/ai-glossary/ai-safety)

---
*From the SPIDITS AI Knowledge Base (336+ terms indexed). Explore full technical definitions: [https://spidits.com/ai-glossary](https://spidits.com/ai-glossary)*