Building an AI Fact-Checker Android Application with Firebase Backend

Abstract

In the era of technological outbreak and far-reaching communication, misinformation has been an ever-concerning matter that requires constant attention from both the government and the media. In several developing countries, the problem appears to be even more serious as cyber-criminals are able to leverage technology in a sophisticated manner to deceive the public or part of the population. The objective of the thesis was to offer a potential solution to that problem by developing an Android application that utilizes a large language model (LLM) to verify the reliability of a given piece of information. The model must have access to a vast amount of real-time information. This will allow it to effectively handle the complexity and comprehensiveness of the application’s functionality. The thesis introduces different tools and technologies that were used to develop the application, such as Android Studio, Jetpack Compose, Gradle, Firebase, and xAI API. It subsequently describes the process of combining these building blocks to construct the usable application. The outcome of thesis indicated that the final product is not yet ready for industrial deployment, but an MVP is able to showcase basic functionalities and the main ideas for its future version.