| Core question answered | Is this claim true? Each claim scored 0-100 with evidence | Was this text written by AI, a human, or a mix? |
| Video, audio & podcast input | Paste a YouTube, TikTok, Instagram Reel or podcast link; or run the extension live while you watch | Text only — accepts pasted text, PDF, DOCX, RTF. Does not process video or audio |
| AI-authorship detection in text | Not its purpose — flags deepfake/AI-media signals, but does not classify text authorship | Purpose-built for it. ~99.98% accuracy and roughly a 1-in-10,000 false-positive rate, independently validated by University of Chicago and University of Maryland researchers |
| Claim-level fact-check with linked sources | Extracts each verifiable claim, cross-references live web sources, links the evidence, flags logical fallacies | Does not fact-check or verify accuracy — authorship only |
| Sentence-level explainability & languages | Explains scores per claim with cited sources | Sentence-level AI signals, human/AI/mixed labels, catches lightly edited and 'humanized' text, 20+ languages |
| Deepfake & synthetic-media signals | Flags deepfake and AI-generated media signals in video and audio | Out of scope — text classifier, not a media forensics tool |