Sign in to analyze
Social Media AI Detection Platform • Deepfake & Manipulation Analysis
v11.7 Powered by PixelForce™ ML Engine • 100 AI Models • 200+ Features • ONNX RuntimeLatest updates, features, and announcements
Analyze videos for AI generation, deepfakes, manipulation, and frame splicing using 100 detection algorithms.
Drop video here or click to select
Supports MP4, MOV, AVI, WebM (max 500MB)
Upload labeled images to improve detection accuracy. The more examples you provide, the better the system learns to distinguish real from fake.
Select 1-2 categories above, then drop images here
You can select multiple categories (e.g., AI + Manipulated)
Upload labeled videos or provide video URLs to train the system on video authenticity patterns.
Drop videos here or click to select
Supports MP4, MOV, AVI, WebM (max 500MB each)
Digital certificates proving your media has been verified by Verifaux™. Share on social media to build trust and combat misinformation.
Analyze an image or video first, then generate your Authentication Card from the results.
An Authentication Card is a shareable digital certificate that proves your image or video has been analyzed and verified by Verifaux™'s 100 AI detection algorithms.
Your card design is determined by the verification results.
Authentication Cards are designed to be shared alongside your content to prove authenticity.
Authentication Cards benefit anyone who needs to prove their content is genuine.
Auth Cards from your recent analyses will appear here.
The Verifaux™ Dev Team aims to build the world's best Social Media AI detection platform. We combine 100 detection algorithms with machine learning to create a comprehensive system for detecting deepfakes, AI-generated images, and manipulated media. Every image and video you analyze makes the system smarter.
Verifaux™ uses a multi-layered detection approach combining rule-based forensic analysis with machine learning. Each image passes through 100 specialized AI detectors, and the results are weighted and combined to produce an authenticity score.
The foundation of Verifaux™'s detection capabilities. These forensic algorithms analyze the fundamental properties of images to detect manipulation at the pixel level.
ELA works by re-compressing an image and analyzing the difference. Manipulated areas show different error levels than the original content.
Examines hidden image data for signs of editing, software manipulation, or inconsistent device information.
Specialized analysis of JPEG compression artifacts, quantization tables, and block boundaries.
Sophisticated algorithms that analyze texture patterns, frequency domains, and statistical distributions to detect subtle manipulation and AI generation.
Mathematical analysis of pixel value distributions to identify anomalies invisible to the human eye.
Fourier transform analysis reveals periodic patterns and artifacts characteristic of AI generation and manipulation.
Analyzes surface textures for signs of smoothing, AI generation artifacts, and unnatural patterns.
Social media influencers commonly use filters and editing apps to enhance their appearance. Verifaux™ detects these modifications and reports exactly what has been altered.
Detection of platform-specific beauty filters from popular social apps.
Detection of dedicated beauty and editing applications.
Individual modifications that Verifaux™ can identify.
AI image generators often struggle with human anatomy. Verifaux™ analyzes facial features, body proportions, and especially hands to detect AI-generated humans.
Examines facial features for biological accuracy and AI generation artifacts.
AI models notoriously struggle with hands. Our specialized detectors examine finger count, joint positions, and nail anatomy.
Full-body analysis for anatomical accuracy and consistency.
Verifaux™ analyzes videos frame-by-frame using 100 specialized algorithms to detect AI-generated videos, deepfakes, frame splicing, and audio manipulation.
Identifies videos generated by AI models like Sora, Runway, Pika, and Kling.
Specialized analysis for face-swap and face-reenactment deepfakes.
Frame-to-frame consistency checks for splicing and manipulation.
Audio track analysis for synthetic voice and manipulation detection.
Verifaux™ is powered by PIXELFORCE™, our advanced machine learning system designed for Social Media AI detection. With 100 specialized neural network models, 200+ extracted features, and ONNX Runtime with GPU acceleration working together, we detect deepfakes, AI-generated images, and manipulated media with exceptional accuracy. The system includes Smart Deduplication to prevent database manipulation and continuously improves through our anti-manipulation learning design.
Specialized neural networks trained for specific detection tasks across 7 categories, working together through ensemble voting.
All 100 ML models are exported to ONNX format for blazing-fast inference with GPU acceleration.
Advanced protection against ML database manipulation through multi-hash duplicate detection.
The system uses tiered learning rates designed to prevent gaming while maximizing accuracy from admin training.
Admin-trained samples provide the highest accuracy boost. This is the main learning source.
Each loaded ONNX model contributes to overall system accuracy through ensemble predictions.
User analyses contribute small incremental improvements. Low rate prevents manipulation.
This ensures admin training always takes priority over user analysis. System cannot be gamed.
Training data is stored in JSON format for easy export/import between local training and production servers.
Each image is analyzed for 200+ features organized into categories for ML training.
Manual training provides high-confidence labeled data for ML model training.
All 100 models vote together for maximum accuracy. Votes are weighted by each model's individual training accuracy.
Verifaux™ Authentication Cards are digital certificates that prove your images and videos have been verified as authentic. Share them on social media to build trust and credibility with your audience.
Authentication Cards are shareable digital certificates generated after Verifaux™ analyzes your media. They display your authenticity score, verification details, and a unique verification code.
Authentication Cards help content creators, journalists, and businesses prove their media is genuine.
Different card designs based on your verification results.
Each Authentication Card includes security features to prevent forgery.
Who benefits from Authentication Cards?
Getting your Authentication Card is simple.
Verifaux™ offers enterprise-grade AI/ML solutions for developers and companies. Access our PixelForce™ AI Detection Algorithms through our Developer API, or license our complete ML Datasets to train your own AI systems.
Full Access to PixelForce™ AI ML System
License PixelForce™ AI Training Data
Thank you for your feedback!
