Date of Award
12-2024
Document Type
Thesis
Degree Name
Master of Science in Information Systems and Technology
Department
Information and Decision Sciences
First Reader/Committee Chair
Oluwatosin Ogundare (Ph.D)
Abstract
ABSTRACT
Generative AI (GenAI) has become a fundamental part of modern life, influencing how we work, learn, and interact with technology. This project focuses specifically on text-based GenAI, which is widely used for tasks such as information gathering, code improvement, and content creation. Despite its benefits, it presents significant security risks that are often underestimated by users. This project investigates these risks and the corporate security gaps that lead to unintentional data leaks. The project also provides a brief overview of Large Language Models (LLMs), which are based on the deep learning technique known as Transformer architecture, used for performing Natural Language Processing (NLP) tasks. Additionally, a brief explanation of the Transformer architecture, including its main components, the encoder and decoder, has been included. The research objectives of this project are: (RO1) to understand how security risks emerge from text-based GenAI and (RO2) to examine how corporate security weaknesses enable unintended data disclosures. Two case studies were selected and analyzed to answer the research objectives. Findings reveal that GenAI safeguards are insufficient, allowing for prompt manipulation and various malicious activities, while inadequate employee training and guidelines contribute to data leaks. Conclusion highlights that, while text-based GenAI offers numerous benefits, it also carries significant security risks. Enhanced corporate training, stronger GenAI security measures, and effective mitigation strategies are essential to address these vulnerabilities. Future research could concentrate on developing these mitigation strategies, establishing standardized guidelines for secure AI usage in corporate environments, and exploring security measures for media-based and audio-based GenAI.
Recommended Citation
Byreddy, Tashya Rakshana, "AN ANALYSIS OF SECURITY RISKS POSED BY TEXT-BASED GENERATIVE AI AND CORPORATE SECURITY WEAKNESSES LEADING TO DATA LEAKS" (2024). Electronic Theses, Projects, and Dissertations. 2087.
https://scholarworks.lib.csusb.edu/etd/2087
Included in
Business Intelligence Commons, Communication Technology and New Media Commons, Computer Engineering Commons, Educational Technology Commons, Engineering Education Commons, Risk Analysis Commons, Systems Science Commons