Every day, millions of images circulate online stripped of their history. A photo of a protest, a portrait of a political figure, a news image from a war zone — you see it, but you have no idea where it came from, whether it was edited, or whether it was generated by AI at all. That gap between image and truth is exactly where misinformation thrives.
Origin Lens is an open source iOS app built to close that gap. It uses C2PA (Coalition for Content Provenance and Authenticity) Content Credentials, EXIF metadata analysis, and SynthID detection to verify image authenticity — entirely on your device, with no data sent to any server.
The Problem: Images Without a Paper Trail
When a photograph is taken with a modern camera or editing tool that supports C2PA, it can be embedded with a cryptographically signed manifest: a tamper-evident record of who created it, what tools were used, and every edit applied. Think of it as a notarized chain of custody for digital media.
The Content Authenticity Initiative, backed by Adobe, the BBC, Canon, Nikon, Leica, and dozens of others, has been pushing this standard into mainstream tools. Major AI generators like DALL-E, Adobe Firefly, and Stable Diffusion already embed C2PA assertions in their outputs. The standard is gaining real traction — but there was no accessible mobile tool to read these credentials in the field.
Origin Lens fills that gap.
What Origin Lens Does
The app analyzes any image you point it at and surfaces:
- C2PA Content Credential verification — validates cryptographic signatures, checks certificate chains, flags tampered content, and identifies signers with timestamps
- AI generation detection — combines C2PA assertions, EXIF metadata, and SynthID signatures to identify AI-generated content from 50+ generators, including Midjourney, DALL-E, Stable Diffusion, and Adobe Firefly
- Complete edit history timeline — shows every capture, edit, and publish action recorded in the manifest
- EXIF metadata analysis — camera make and model, software used, timestamps, artist attribution, and AI generation markers
- Optional reverse image search — searches Bing, Yandex, and Google to detect out-of-context images (with a clear privacy warning before any upload)
You can analyze images from your photo library, from iCloud/Files, or directly from a URL — no download required.
Privacy First
The core verification pipeline runs entirely on-device. There is no analytics, no tracking, no account required, and no server involved in the C2PA or EXIF analysis. The optional reverse image search is the only feature that touches external services, and the app shows an explicit privacy notice before doing so. You can also bring your own API keys.
This matters. A verification tool that leaks the images you are checking would be worse than useless.
The Research Behind It
Origin Lens is more than an app — it is also the subject of a peer-reviewed paper accepted at the ACM Web Conference 2026 (WWW ’26 Companion) in Dubai:
Origin Lens: A Privacy-First Mobile Framework for Cryptographic Image Provenance and AI Detection
Alexander Loth, Dominique Conceicao Rosario, Peter Ebinger, Martin Kappes, Marc-Oliver Pahl
arXiv:2602.03423
The paper details the architecture, the C2PA verification pipeline, and the multi-source AI detection approach. It is part of a broader research initiative on generative AI and misinformation that also includes JudgeGPT and the verification-crisis expert survey.
Open Source
Origin Lens is open source under the GNU General Public License v3.0. The full source code, build instructions, and architecture documentation are available on GitHub.
Get It
Origin Lens is available for iPhone and iPad on the App Store:
Download Origin Lens on the App Store
If you are a researcher, journalist, or fact-checker working on media verification, I would love to hear how you use it. File issues, suggest features, or contribute on GitHub.
