Advanced AI-powered analysis to identify manipulated media with state-of-the-art deepfake detection technology. Built to safeguard digital integrity in the era of synthetic media.

Deepfake Detection System
employs a consensus-based multi-judge analysis system where six specialized AI judges independently evaluate media for signs of manipulation. This approach eliminates bias and provides unmatched detection accuracy. Hopefully, the next update will make it 99.9% perfect. Every major feature that isn’t available right now will be added in the next few updates, including the features already listed. Due to temporary backend limitations, the detection page may occasionally become unavailable if my laptop goes offline or loses internet connection. A proper dedicated backend will be deployed within the next few days.
In an era where AI-generated deepfakes threaten information integrity, SOTTO stands as Bangladesh's first line of defense. Created with the mission to protect digital truth and empower investigators with tools to combat synthetic media.
Confirm the authenticity of critical media with confidence using our advanced analysis system.
Uncover even the most convincing AI-generated media with our multi-judge system.
Analyzes facial landmarks, biometric inconsistencies, and identity mismatches across frames.
Detects GAN signatures, neural rendering artifacts, and synthetic image generation patterns.
Validates semantic consistency, metadata integrity, and source authenticity.
Examines lighting direction, reflections, shadows, and environmental realism inconsistencies.
Combines evidence from all judges and delivers the final authoritative verdict.
Detects SynthID watermarks and advanced AI-generation fingerprints in media files.
Each judge independently analyzes the media.
Judges provide binary verdicts with confidence scores.
The Chief Justice synthesizes all evidence and delivers the final authoritative verdict
Consensus is reached by majority vote, with the Chief Justice having final authority
Eliminates Bias: Multiple independent analyses prevent single-point failures
Specialized Expertise: Each judge focuses on specific forensic domains
Higher Accuracy: Consensus reduces false positives and false negatives
Transparent Process: Users see reasoning from each judge for full transparency
SOTTO leverages cutting-edge AI models with advanced multimodal understanding, superior reasoning capabilities, and the ability to analyze both images and videos with unprecedented accuracy. The system's deep context analysis allows for comprehensive examination of complex media.
JUNAID KABIR @ LUCIFER
Pioneer Bangladeshi developer and AI expert who created SOTTO - Bangladesh's first deepfake detection system. Combining expertise in artificial intelligence, computer vision, and forensic analysis to combat the growing threat of synthetic media.
By using SOTTO, you agree that your uploaded media will be processed through our multi-model AI system for deepfake detection. Your images may be temporarily stored on our secure servers and could be used to train our models to improve detection accuracy.
We recommend avoiding sharing personal or sensitive images. Data stored in our system is kept for a limited amount of time and is handled in accordance with our privacy policy and terms of service.