System prompt leakage
When a bot reveals instructions or information that its application kept hidden from the user.
The fuller definition
Disclosure of system instructions or other hidden context supplied to a model.
- Context
Hidden material is supplied
- Disclosure
The response reveals it
- Control
Keep secrets outside model context
A simplified relationship, not a screenshot or a complete design.
A concrete example
A response reveals an internal instruction or a sensitive value included in the prompt.
The distinction that matters
Instructions should not contain secrets or serve as access controls. The consequences depend on what is disclosed.
Related terms
System instruction
Standing instructions the application gives the model before your conversation starts, such as how to behave or what to refuse. Also called a system prompt.
Reliability & securityPrompt injection
Text in a source that tries to redirect an AI tool. It may tell the tool to ignore your rules or leak data.
Reliability & securityData leakage
Data reaching a place it should not. In model tests, it can also mean test answers have reached the training data.
Where it is taught
Test an AI workflow on work you can assessSource and further reading
OWASP GenAI · LLM07:2025 System Prompt LeakageThe definition and example are written for this library. The link gives background, official documentation or legal authority as labelled. Legal scope depends on the jurisdiction and current text; product features depend on release and settings.