Seeing through AI-generated images.
Lens analyses every image on any webpage and tells you whether it was AI-generated — right where you browse. No uploads. No account. No tracking.
Four layers of detection
How it works
When you visit a page, Lens quietly checks every image in the background. It runs up to four detection passes, stopping as soon as it has enough confidence.
1. URL & filename recognition
Recognises origin patterns from OpenAI, Midjourney, Adobe Firefly, Stability AI, xAI, Bing Image Creator, and a dozen other services — a dead giveaway before even looking at pixels.
2. C2PA metadata
Many AI tools embed machine-readable provenance using the C2PA Content Credentials standard. ChatGPT, Adobe Firefly, and Bing Image Creator all do this. Lens reads these credentials directly.
3. Pixel statistics
When metadata has been stripped, Lens analyses noise patterns, gradient smoothness, and inter-block noise correlation. Real photographs have distinctive sensor fingerprints. AI images don’t.
4. Frequency watermarks
Some AI systems embed invisible watermarks using spread-spectrum techniques. Lens uses FFT-based analysis to detect these, including signals consistent with Google SynthID.
What you see
Results appear as a small badge overlaid on each image. Hover to see exactly what was detected, the confidence weight, and which detection layer caught it.
Privacy by design
No uploads
Images are analysed in your browser. Nothing is sent to any server — ever.
No account
No email. No sign-up. Install and it works — that’s it.
No telemetry
No analytics, no tracking, no data brokers. Image bytes are discarded immediately after analysis.
Open source
Every line of detection logic is on GitHub. Audit it, verify our claims, or contribute.
Fully open source · MIT licensed
Source code on GitHub. Read it, audit it, contribute.