The Forensic Pipeline V2

Discover exactly how Truth Lenses detects digital manipulation - from client-side metadata extraction to high-speed multimodal vision verification.

01

Client Metadata & In-Memory Sampling

When you analyze media, it never touches persistent storage. The browser extracts EXIF camera metadata and inspects for C2PA provenance markers locally. Large images (>2048px) are smoothly downscaled on Canvas, while video clips are sampled locally into 8 key frames-raw video files never leave your device.

02

Forensic Parameterization

We supply a structured system prompt instructing the vision model to inspect visual clues in plain English: anatomical inconsistencies, lighting irregularities, texture smoothing, and temporal flickering across video frames-grounded by client metadata context.

03

Stateless Multi-Model Free Cascade

The media payload is analyzed statelessly in memory using a resilient priority cascade across Google Gemini models (Gemini 4 Hydrogen, Gemini 3.8 Flash, Gemini 3.6 Flash, Gemini 3.5 Flash-Lite). The API checks structural patterns and discards all data immediately upon completion.

04

Rich Structured Verdict

The analysis returns a validated JSON schema containing a verdict (REAL, FAKE, ALTERED, UNCERTAIN), confidence percentage (0–100), categorized clue cards with severity ratings, file provenance context, and practical next steps.

How Truth Lenses Works

See how Truth Lenses samples media in your browser and runs stateless AI visual checks with zero file storage.

Verdicts, Reliability & Honest Limits

What each result means and how the pipeline stays accurate and private

What REAL & FAKE Mean

REAL indicates consistent photographic lighting, natural textures, and expected lens cues. FAKE indicates strong patterns of full AI generation or synthetic face-swapping.

What ALTERED & UNCERTAIN Mean

ALTERED flags genuine media with targeted edits, composites, or heavy retouching. UNCERTAIN appears when heavy compression or low detail prevents a clear call.

Browser-First Preprocessing

Oversized photos resize in-browser, EXIF and C2PA markers parse locally via exifr, and video files stay on-device while Canvas samples 8 preview frames.

Multi-Model Gemini Cascade

Requests start on gemini-4-hydrogen and automatically fail over across Gemini Flash models during demand spikes for fast, resilient results.

No Watermark or Filename Bias

Our prompt instructs the model not to treat corner blog logos, publisher stamps, or stripped WebP metadata as proof that an image is real.

Probabilistic Verification Only

No automated detector is 100% infallible. Always pair tool verdicts with reverse-image search and source checks for high-stakes decisions.

Technical & Privacy FAQs

Photos resize locally if over 2048px while EXIF/C2PA markers are read in your browser; videos extract 8 JPEG frames locally. Only that stateless payload enters server memory for Gemini inspection and is discarded immediately afterward.
FAKE is used when an image or scene appears predominantly AI-generated or face-swapped, whereas ALTERED points to localized editing, compositing, or manipulation inside an otherwise real frame.
Single AI endpoints can slow down or hit demand limits. Cascading from gemini-4-hydrogen through Gemini Flash fallback models keeps response times fast and reliable.
Both real photos and AI-generated blog graphics lose EXIF metadata on WebP export and often carry corner logos. Truth Lenses inspects underlying lighting, geometry, and textures independently.
Run a reverse-image search (Google Lens or TinEye) to find earlier appearances, inspect the original uploader's history, and verify claims against trusted primary sources.

Image vs. Video Pipeline at a Glance

Truth Lenses uses distinct browser-side preprocessing tailored to each media format before submitting payloads to the stateless Gemini vision cascade.

Image Forensic Checker

Single-frame high-resolution visual and metadata analysis

Supported Formats
JPG, PNG, WEBP, or direct Image URL
Browser Preprocessing
On-device EXIF & C2PA marker scan via exifr; auto Canvas resize if >2048px
Daily Quota
3 free guest scans / day (no login required)
Inspection Scope
Full pixel geometry, specular lighting, textures, camera metadata, and C2PA provenance
Scope Note
Heavily compressed social media reposts may lose camera EXIF

Video Deepfake Scanner

Multi-frame sequential analysis with on-device extraction

Supported Formats
MP4, MOV, WEBM (up to 30 MB)
Browser Preprocessing
HTML5 Canvas extracts 8 key frames locally in browser; raw video never leaves device
Daily Quota
3 free scans / day (free Google Sign-In required)
Inspection Scope
Visual frame consistency, face-swap blending borders, warping, and AI motion artifacts
Scope Note
Visual frames only (no audio/speech track inspection yet)

Zero Storage Guarantee on Both Pipelines

Whether scanning a photo or a video clip, neither raw media files nor sampled frames are ever saved to disk, stored in databases, or used to train AI models. In-memory payloads are permanently discarded immediately after your verdict is rendered.

Read our Privacy Policy