PsyScam: A Benchmark for Psychological Techniques in Real-World Scams

TL;DR

PsyScam is a benchmark dataset that captures psychological techniques used in real-world scams and evaluates how well models can detect and generate scam content leveraging these techniques, highlighting challenges for current AI systems.

Contribution

Introduces PsyScam, a comprehensive benchmark dataset with annotations of psychological techniques in scams, and explores LLMs' ability to generate scam variants based on these techniques.

Findings

Existing models struggle to detect scams using PTs.

LLMs can generate diverse scam variants based on PTs.

PsyScam challenges current AI defenses against scams.

Abstract

Over the years, online scams have grown dramatically, with nearly 50% of global consumers encountering scam attempts each week. These scams cause not only significant financial losses to individuals and businesses, but also lasting psychological trauma, largely due to scammers' strategic employment of psychological techniques (PTs) to manipulate victims. Meanwhile, scammers continually evolve their tactics by leveraging advances in Large Language Models (LLMs) to generate diverse scam variants that easily bypass existing defenses. To address this pressing problem, we introduce PsyScam, a benchmark designed to systematically capture the PTs employed in real-world scam reports, and investigate how LLMs can be utilized to generate variants of scams based on the PTs and the contexts provided by these scams. Specifically, we collect a wide range of scam reports and ground its annotations of employed PTs in well-established cognitive and psychological theories. We further demonstrate LLMs' capabilities in generating through two downstream tasks: scam completion, and scam augmentation. Experimental results show that PsyScam presents significant challenges to existing models in both detecting and generating scam content based on the PTs used by real-world scammers. Our code and dataset are available.