CITATION — REFERENCE ENTRY
data-instruction-blur · greshake-2023-indirect-prompt-injection
- Claim ID
- data-instruction-blur
- Assertion
- Greshake and colleagues argue that LLM-integrated applications blur the line between data and instructions, allowing adversaries to remotely affect users' systems by strategically injecting prompts into data that the model is likely to retrieve at inference time.
- Quote
We argue that LLM-Integrated Applications blur the line between data and instructions.
- Quote language
- en
- Locator
- page: 1
Available in