Large Language Models (LLMs) have gained prominence in various applications, including security. This paper explores the utility of LLMs in scam detection, a critical aspect of cybersecurity. Unlike traditional applications, we propose a novel use case for LLMs to identify scams, such as phishing, advance fee fraud, and romance scams. We present notable security applications of LLMs and discuss the unique challenges posed by scams. Specifically, we outline the key steps involved in building an effective scam detector using LLMs, emphasizing data collection, preprocessing, model selection, training, and integration into target systems. Additionally, we conduct a preliminary evaluation using GPT-3.5 and GPT-4 on a duplicated email, highlighting their proficiency in identifying common signs of phishing or scam emails. The results demonstrate the models' effectiveness in recognizing suspicious elements, but we emphasize the need for a comprehensive assessment across various language tasks. The paper concludes by underlining the importance of ongoing refinement and collaboration with cybersecurity experts to adapt to evolving threats.
ScamSweeper: Detecting Illegal Accounts in Web3 Scams via Transactions Analysis
Hainan University（海南大学）；Ohio State University（俄亥俄州立大学）；College of Computer and Cyber Science, Augusta University（奥古斯塔大学计算机与网络科学学院）；Jinan University（暨南大学）；Keen Security Lab, Tencent（腾讯科恩实验室）；University of Chinese Academy of Sciences（中国科学院大学）
The prosperity of the cryptocurrency ecosystem drives the need for digital asset trading platforms. Beyond centralized exchanges (CEXs), decentralized exchanges (DEXs) are introduced to allow users to trade cryptocurrency without transferring the custody of their digital assets to the middlemen, thus eliminating the security and privacy issues of traditional CEX. Uniswap, as the most prominent cryptocurrency DEX, is continuing to attract scammers, with fraudulent cryptocurrencies flooding in the ecosystem. In this paper, we take the first step to detect and characterize scam tokens on Uniswap.…
Soon after its introduction in 2009, Bitcoin has been adopted by cyber-criminals, which rely on its pseudonymity to implement virtually untraceable scams. One of the typical scams that operate on Bitcoin are the so-called Ponzi schemes. These are fraudulent investments which repay users with the funds invested by new users that join the scheme, and implode when it is no longer possible to find new investments. Despite being illegal in many countries, Ponzi schemes are now proliferating on Bitcoin, and they keep alluring new victims, who are plundered of millions of dollars. We apply data minin…
Nokia Bell-labs（诺基亚贝尔实验室）；Universitat Pompeu Fabra（庞培法布拉大学）；Universitat de Barcelona（巴塞罗那大学）
Uniswap, like other DEXs, has gained much attention this year because it is a non-custodial and publicly verifiable exchange that allows users to trade digital assets without trusted third parties. However, its simplicity and lack of regulation also makes it easy to execute initial coin offering scams by listing non-valuable tokens. This method of performing scams is known as rug pull, a phenomenon that already existed in traditional finance but has become more relevant in DeFi. Various projects such as [34,37] have contributed to detecting rug pulls in EVM compatible chains. However, the firs…
Love, Lies, and Language Models: Investigating AI's Role in Romance-Baiting Scams
Center for Cybersecurity Systems & Networks, Amrita Vishwa Vidyapeetham, Amritapuri（网络安全系统与网络中心，阿米特大学，阿米塔普里）；Ca’ Foscari University of Venice（威尼斯卡弗斯卡里大学）；University of Melbourne（墨尔本大学）；Ben Gurion University of the Negev（内盖夫本· Gurion大学）
Exposing LLM Vulnerabilities: Adversarial Scam Detection and Performance
Department of Computer Science, Virginia Tech, USA（维吉尼亚理工学院计算机科学系）；Department of Information Sciences（信息科学系）；School of Policy and Government, George Mason University, USA（乔治·马歇尔大学政策与政府学院）；School of Social Work, Indiana University, USA（印第安纳大学社会工作学院）
Recently, deep learning techniques have garnered substantial attention for their ability to identify vulnerable code patterns accurately. However, current state-of-the-art deep learning models, such as Convolutional Neural Networks (CNN), and Long Short-Term Memories (LSTMs) require substantial computational resources. This results in a level of overhead that makes their implementation unfeasible for deployment in realtime settings. This study presents a novel transformer-based vulnerability detection framework, referred to as VulDetect, which is achieved through the fine-tuning of a pre-train…
Cyber attacks such as phishing, IRS scams, etc., still are successful in fooling Internet users. Users are the last line of defense against these attacks since attackers seem to always find a way to bypass security systems. Understanding users' reason about the scams and frauds can help security providers to improve users security hygiene practices. In this work, we study the users' reasoning and the effectiveness of several variables within the context of the company representative fraud. Some of the variables that we study are: 1) the effect of using LinkedIn as a medium for delivering the p…
LASIGE, Faculdade de Ciências, Universidade de Lisboa（里斯本大学理学院 LASIGE 实验室）；INESC-ID, Instituto Superior Técnico, Universidade de Lisboa（里斯本大学高等技术学院 INESC-ID 研究所）
Web applications continue to be a favorite target for hackers due to a combination of wide adoption and rapid deployment cycles, which often lead to the introduction of high impact vulnerabilities. Static analysis tools are important to search for bugs automatically in the program source code, supporting developers on their removal. However, building these tools requires programming the knowledge on how to discover the vulnerabilities. This paper presents an alternative approach in which tools learn to detect flaws automatically by resorting to artificial intelligence concepts, more concretely…
End-to-end (E2E) models are often being accompanied by language models (LMs) via shallow fusion for boosting their overall quality as well as recognition of rare words. At the same time, several prior works show that LMs are susceptible to unintentionally memorizing rare or unique sequences in the training data. In this work, we design a framework for detecting memorization of random textual sequences (which we call canaries) in the LM training data when one has only black-box (query) access to LM-fused speech recognizer, as opposed to direct access to the LM. On a production-grade Conformer R…
