Search and Filter

Synthguard: An Open Platform For Detecting Ai-Generated Multimedia With Multimodal Llms

Date

AI Overview

Why it matters

AI-generated media can spread misinformation and erode public trust. Existing deepfake detection tools are often closed-source, limited in modality, and lack transparency, making it hard for users to understand how detection decisions are made.

Study at a glance
  • Question: Can an open platform effectively combine traditional detectors and multimodal models to provide explainable deepfake detection across media?
  • Setting and data: Purdue University Linux server hosting SynthGuard, evaluated using pre-trained models on the AI-Face benchmark.
  • Method: System development and prototype demonstration of a modular full-stack web platform.
  • AI tools: SynthGuard, Qwen-VL-Chat, InternVL-Chat-V1.5, LLaVA-NeXT, Whisper, Qwen2-VL-2B
Key findings
  • SynthGuard provides a publicly accessible platform that unifies detection for both AI-generated image and audio content in one system.
  • The platform integrates multimodal language models to supply natural-language descriptions of image attributes and audio characteristics rather than binary authenticity labels alone.
  • Unlike existing tools that focus primarily on single modalities without explanation, the system combines specialized forensic detectors, fairness-enhanced models, and conversational assistance.

Summary drafted with AI from the full paper and may contain errors. See the full paper for details.

Artificial Intelligence (AI) has made it possible for anyone to create images, audio, and video with unprecedented ease, enriching education, communication, and creative expression. At the same time, the rapid rise of AI-generated media has introduced serious risks, including misinformation, identity misuse, and the erosion of public trust as synthetic content becomes increasingly indistinguishable from real media. Although deepfake detection has advanced, many existing tools remain closed-source, limited in modality, or lacking transparency and educational value, making it difficult for users to understand how detection decisions are made. To address these gaps, we introduce SynthGuard, an open, user-friendly platform for detecting and analyzing AI-generated multimedia using both traditional detectors and multimodal large language models (MLLMs). SynthGuard provides explainable inference, unified image and audio support, and an interactive interface designed to make forensic analysis accessible to researchers, educators, and the public. The SynthGuard platform is available at: https://in-engr-nova.it.purdue.edu/

Research Categories

What is the application?
Who is the user?
Who age?
Why use AI?

Publication Source

Publisher
arXiv