most citedEvaluating Large Language Models' Capability to Launch Fully Automated Spear Phishing Campaigns: Validated on Human Subjects

5 citations · 6 across the 3 of their papers we have counts for

collaborators
Showing cs.AIShow all

1 paper · 1 filter

cs.AI2025

Deceptive Automated Interpretability: Language Models Coordinating to Fool Oversight Systems

Simon Lermen, Mateusz Dziemian, Natalia Pérez-Campanero Antolín

We demonstrate how AI agents can coordinate to deceive oversight systems using automated interpretability of neural networks. Using sparse autoencoders (SAEs) as our experimental f…