Jailbreak
[/ˈdʒeɪlbreɪk/]
Definitions
A technique used to bypass the safety filters and content policies of an AI model, typically by framing harmful requests in ways the model's defenses don't recognize. Jailbreaks often use role-play scenarios, hypothetical framings, or encoded instructions to make the model comply with prohibited requests.
“The "DAN" jailbreak asked the model to pretend it was an AI with no restrictions.”
Related Terms
- AI AlignmentAI & Technology
The research field focused on ensuring that AI systems pursue goals that match human values and intentions. A misaligned AI might optimize for a metric that appears correct but pro...
- Constitutional AIAI & Technology
A training methodology developed by Anthropic where a set of guiding principles (a "constitution") is used to self-supervise and refine AI outputs. The model critiques and rewrites...
- GuardrailsAI & Technology
Safety constraints and filters applied to AI systems to prevent harmful, offensive, or out-of-scope outputs. Guardrails can be implemented at the model level (via training), prompt...
- Prompt InjectionAI & Technology
A security attack where malicious instructions are embedded in user-provided input to override or hijack an AI system's intended behavior. Analogous to SQL injection, prompt inject...
- AgenticAI & Technology
Describing AI systems capable of autonomous action, planning, and decision-making. An agentic AI can break down tasks, use tools, and work toward goals with minimal human intervent...
- Chain of ThoughtAI & Technology
A prompting technique where a language model is encouraged or required to show its step-by-step reasoning before providing a final answer. Chain-of-thought prompting significantly ...