prompt injection
/ˈprɒmpt ɪnˌdʒekʃən/
an attack or failure mode in which instructions embedded in user input or external content steer an LLM application away from its intended behavior
prompt injection in a sentence
“A webpage contained an indirect prompt injection telling the browsing agent to disclose private data.”
Origin of prompt injection
Latin promptus + injectio a throwing in
Related Words
context priming
using early context to set expectations and influence subsequent model behavior
meta-prompting
asking the model to help design or improve prompts for itself
persona prompting
instructing a model to adopt a role or perspective to shape its approach, emphasis, or style
instruction tuning
fine-tuning models specifically on instruction-following examples
structured output
model output constrained to follow a defined machine-readable schema, commonly a supported subset of JSON Schema
delimiter
a character or tag that marks where one part of a prompt ends and another begins