Integrating Generative AI with Context-Aware Multi-Factor Authentication to Enhance BYOD Security

Traditional Context-Aware Multi-Factor Authentication (CAMFA) algorithm-based software for authenticating personal devices can't assess how different environments affect Bring Your Own Device (BYOD) rules and policies. This limitation arises from their inability to effectively process multimedia data, which is essential for representing different types of environments. To address this, we have developed software that utilizes Retrieval-Augmented Generation (RAG) to customize Large Language Models (LLMs) for the CAMFA algorithm, enabling it to effectively process multimedia data for BYOD authentication. This new CAMFA algorithm-based BYOD authentication software enables intelligent authentication decisions based on the combination of text, audio, and image data, capturing multiple contextual factors of different environments, such as location, device type, user behavior, and environmental dynamics. This paper presents the design and development of this software, as well as the testing results. Our work provides an effective approach for organizations seeking to strengthen BYOD security by leveraging the new capabilities offered by Generative AI (GAI) and LLMs.