RESEARCHInvestigateNEXT 12 MONTHS
When Agents Learn to Be You: Benchmarking Privacy Leakage, Impersonation Risk, and Defenses in Persona Skills
arXiv cs.CL — Computation and Language
Factual evidence
What the source reports
Researchers introduce AntiSkillBench to evaluate privacy leakage and impersonation risks in agentic AI systems utilizing persona skills.
Open source