Analysis of 14,727 security and privacy prompts from WildChat finds commercial LLMs give higher-quality responses than open-weight models but can produce inconsistent answers across repeated queries.
and Ippolito, Daphne and Tram
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
Formalizes agentic surveillance, releases SurveilBench for testing AI reporting behaviors across corporate, education, and police scenarios, and develops three prompt-injection evasion techniques.
AI agents enable adaptive computer worms that propagate autonomously by reasoning about targets and synthesizing attacks using LLMs on stolen compute.
citing papers explorer
-
Security and Privacy Prompts in the Wild: What Users Ask LLMs and How LLMs Respond
Analysis of 14,727 security and privacy prompts from WildChat finds commercial LLMs give higher-quality responses than open-weight models but can produce inconsistent answers across repeated queries.
-
AI Snitches Get Glitches: Towards Evading Agentic Surveillance
Formalizes agentic surveillance, releases SurveilBench for testing AI reporting behaviors across corporate, education, and police scenarios, and develops three prompt-injection evasion techniques.
-
AI Agents Enable Adaptive Computer Worms
AI agents enable adaptive computer worms that propagate autonomously by reasoning about targets and synthesizing attacks using LLMs on stolen compute.