Protecting a source who sends you photos

A source sends you a photo of a document, a meeting room or a protest. Before you publish it, or even forward it to an editor, the file itself may say more about the sender than the picture does. This article covers what is in the file and how channels treat it. It is not legal advice: rules on protecting sources differ from country to country, so ask your newsroom’s lawyer about yours.

What a photo file can say about the person who took it

Phone photos can carry Exif metadata. In a JPEG, the Library of Congress explains, Exif sits in application marker segments (APP1 and APP2) inside the file. The standard is maintained by JEITA and CIPA; the current version is Exif 3.1, published by CIPA in 2026.

Fields that matter for a source:

• Location. GPS tags such as GPSLatitude, GPSLongitude and GPSAltitude are part of Exif. Apple says that when the Camera app has Location Services access, coordinates “are embedded into each photo and video.”
• Time. DateTimeOriginal records when the photo was taken.
• Device. Make and Model name the phone or camera.
• Software. In a 2025 peer-⁠reviewed test, iPhone photos recorded the iOS version number and Android photos recorded a firmware or build ID. That can narrow down which phone, and which update, the sender has.

Some files also carry C2PA Content Credentials, an open standard “to establish the origin and edits of digital content.” In a JPEG, the C2PA record sits in APP11 marker segments, usually several of them because each segment is limited to 64 KB.

Then there is the picture itself. Faces in the background, a reflection in a window, a name on a screen or a badge can identify someone more directly than any metadata.

How channels treat metadata

There is less official information here than you might expect, and what exists does not always agree.

Read the table with care. The Soni test used 10 images and did not state app versions. Its finding on Signal’s document mode and Signal’s own code point in different directions. Apps change with updates. The safe working rule: assume every channel keeps everything, and clean the file yourself.

One more channel detail affects you, the receiver. Apple says photos shared with you in Messages “are automatically saved in your photo library.” A source’s photo can land next to your own pictures, and in anything that syncs your library. To stop that, go to Settings > Apps > Messages > Shared with You and turn off Photos.

What your source can do before sending

If you can talk to the source first, these steps come from Apple’s own guides:

1. Stop recording location in new photos: Settings > Privacy & Security > Location Services > Camera > Never.
2. Remove location from a photo already taken: open it, tap Info, then Adjust Info > Adjust Location > No Location.
3. Leave location out when sharing: select the photo, tap Share, tap Options, turn off Location. “Location Not Included” then shows above the Options button. Apple documents this in its Personal Safety guide; check it appears on the source’s version of iOS.
4. Leave All Photos Data off in AirDrop.

The Location switch covers location. Apple’s note also lists date, time and device, and it doesn’t say those are dropped, so assume they still travel. A metadata remover clears those too.

What to do before you publish

• Clean the copy you publish. The photo metadata tool linked below shows what a file holds, including any C2PA record, and saves a copy without it.
• Blur what the picture gives away. Use the face blur tool for faces, plates, badges and screens. For text, choose a solid block; a light blur on text can sometimes be reversed.
• Keep the original separately. C2PA’s specification notes that “an asset can become separated from its C2PA Manifest due to removal or corruption of asset metadata.” If you need provenance to verify the image later, keep the untouched original in safe storage and publish only the cleaned copy.

Storing source material on your phone

Your photo library is a convenient place for source material and a poor one. With iCloud Photos on, Apple’s default is standard data protection, where “the encryption keys are secured in Apple data centers.” Advanced Data Protection makes Photos end-⁠to-⁠end encrypted, but Apple says Shared Albums don’t support it. And Apple says the Hidden album unlocks with Face ID, Touch ID or your passcode, so anyone who knows the passcode can open it.

A vault app that encrypts with its own key, separate from your passcode, keeps originals out of the library: out of iCloud Photos, out of Shared Albums, and away from apps with photo access.

What it can’t do:

• Protect the source’s phone. The original, with its metadata, may still be on the sender’s device and in their backups.
• Protect what you share out. Once you send or publish a file, the copy is outside the vault.
• Protect what is on screen, or a phone running malware while the vault is open.

Checklist

• Ask the source to turn off camera location and to send a cleaned file.
• Assume the channel kept all metadata, whichever one they used.
• Turn off Messages > Shared with You > Photos so received photos don’t land in your library.
• Before publishing, strip metadata and blur faces, screens and badges.
• Keep the original apart from your photo library, behind its own key.
• Ask your newsroom’s lawyer about the rules where you work.