by Sightengine
Sightengine is an API for moderating images, video, text and audio and for detecting AI-generated media and deepfakes.
Updated EU status checked Last shown
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
Sightengine is a commercial API service for automated content moderation and media analysis. It covers image and video moderation (150+ classes such as nudity, violence, drugs, weapons and self-harm), text moderation, OCR and QR-code screening, audio moderation, near-duplicate and spam detection, image quality assessment, and profile-image validation.
Its AI content detection products flag AI-generated images, video, speech and music, and detect deepfakes and face swaps. The vendor states that detection of speech and music works from the audio alone, without watermarks or metadata. Identity-related tools include liveness detection, recapture detection (photos of screens or printouts) and age-group estimation.
Access is through a REST API with SDKs and a web demo. Plans are tiered (Free, Starter, Pro, Enterprise); the pricing page lists regional endpoints for processing media in a chosen region, and geo-fencing, custom models and SLAs on the Enterprise tier. In a YouTube test, Matt Wolfe reported that its AI video detection flagged AI videos correctly but returned no explanation for its verdicts and consumed many operations (about 12,000 across roughly six videos), so cost can add up for video. Video upload limits are 50MB on lower tiers, with 500MB and live-stream moderation on Pro.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
A third-party computer-vision API for AI-generated content detection that finally made the detector work. It flagged AI videos correctly, but gives no reasons and is expensive: roughly 12,000 operations were used in testing across about six videos.
Closest neighbours by meaning, not just by category.
App · Zoom
Zoom feature that analyzes video and audio in real time to flag possible deepfake participants, with a confidence indicator.
App
Creator-built app that takes a video URL and returns an AI-generated-or-not verdict using Gemini and SightEngine.
App · Google
Google's verification portal for checking whether images, video, audio or text carry a SynthID AI-content watermark.
Model · DeepSeek
Experimental vision-enabled version of DeepSeek V4 Flash with image understanding; open weights under MIT, now retired on the API.
App · The Atlantic
Searchable Atlantic database showing which music and videos were used to train AI models such as Suno and Udio.
Library · Egonex AI
Claude Code plugin that builds an interactive knowledge graph of a codebase or wiki using a multi-agent pipeline.
