5dot checks video and audio for AI-generated voices, deepfake footage, and lip-sync tampering — plain-language results first, full technical detail when you want it.
Free to start. Automated result, not a guarantee.
Catches synthesized and cloned voices by analyzing spectral and prosodic patterns across short overlapping segments of audio.
Scores visual consistency frame-by-frame to catch face-swaps, mouth-movement mismatches, and GAN artifacts.
Finds cut-and-splice edits, re-encoding jumps, and timeline discontinuities — AI-made or not.
How it works
A file from your device, or a link from YouTube, TikTok, Instagram, Facebook, or X.
Voice, video, lip-sync — run one or all three on the same clip.
A clear result first. Tap through for the exact score and a timeline chart.
A full PDF forensic report for any completed scan, ready to share.
An honest tool, not a magic one. 5dot's detection is powered by machine learning and gives you a strong signal — not absolute proof. Every result ships with a confidence score, and we're upfront that no detector catches everything. Use it to inform your judgment, not replace it.