The increasing accessibility and sophistication of generative AI have enabled the creation of highly realistic deepfakes and other forms of synthetic media, introducing new challenges for fraud detection and digital security. This project investigates the use of computer vision and machine learning techniques to detect fraudulent AI-generated content, with a particular focus on deepfake identification. The research involves evaluating existing detection models, developing and training machine learning pipelines, experimenting with diverse datasets, and analyzing model performance and robustness against increasingly advanced generative methods. By contributing to the development of reliable tools for identifying manipulated media, this work aims to help mitigate deepfake-based fraud, strengthen trust in digital information, and support safer online environments while expanding expertise in computer vision, machine learning, and fraud detection.

I am participating in SuperUROP because it provides an opportunity to contribute to research on a problem with significant real-world impact while deepening my understanding of machine learning and computer vision. As generative AI continues to make deepfakes and other forms of synthetic media more convincing, developing reliable methods to detect fraudulent content has become increasingly important. Through this project, I hope to help advance tools that improve digital trust while gaining valuable research experience working alongside faculty on cutting-edge techniques. SuperUROP will allow me to strengthen both my technical skills and my ability to apply them to meaningful challenges with broad societal relevance.