Systematic testing of ten LLM agents across 20 tool scenarios and 14 attacks finds universal vulnerability to prompt injection enabling data exfiltration, with tooling amplifying leakage.
Using GPT-eliezer against ChatGPT jailbreaking,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Whispers in the Machine: Confidentiality in Agentic Systems
Systematic testing of ten LLM agents across 20 tool scenarios and 14 attacks finds universal vulnerability to prompt injection enabling data exfiltration, with tooling amplifying leakage.