Our project, TruthGuard, an AI-powered platform, is designed to combat misinformation by analyzing digital content for authenticity. It integrates multiple tools into a system comprising a dashboard, home page, AI text detector, deepfake detector, and interactive deepfake game. It also leverages advanced AI models like OpenAI for text analysis and Sightengine for deepfake detection.
The AI text detector, powered by OpenAI, evaluates the statements' accuracy by analyzing their factual correctness and assigning a confidence score supported with explanations. For instance, when asked whether "humans are mammals," it verifies the claim and provides supporting explanations.
Utilizing Sightengine's deepfake detection technology, the deepfake detector examines images to determine their authenticity, enabling users to distinguish between actual and AI-generated content. Leveraging machine learning techniques, it assesses pixel patterns, inconsistencies, and metadata to provide a reliability detection score.
Meanwhile, the deepfake game engages users interactively, allowing them to test their ability to spot manipulated content and improve their digital literacy.
Overall, TruthGuard aims to enhance media transparency by equipping users with AI-driven verification tools, fostering critical thinking, and mitigating the spread of misinformation in an increasingly AI-influenced world.
Log in or sign up for Devpost to join the conversation.