Remove Content Credentials (C2PA) From Images

What Content Credentials are

Content Credentials are a signed record that travels with an image. The record says where the image came from and how it was changed. The open standard behind them is C2PA, from the Coalition for Content Provenance and Authenticity. A tool signs the record, so a viewer can see whether someone changed it afterwards.

Many tools and devices write them:

• Adobe: Adobe says that some of its apps and services embed Content Credentials in the files that they make (Adobe Help Center, 2025).
• OpenAI: OpenAI says that images made with ChatGPT, Codex and its API carry C2PA metadata and SynthID signals (OpenAI Help Center, checked November 2026).
• Google: Google announced that the Pixel 10 camera adds Content Credentials to photos (Google blog, 2025).
• Camera makers: Leica, Sony and Nikon each describe a Content Credentials feature for selected camera models, on their own websites.

So the record is not only for AI images. A photo from a camera can carry one too.

Where the record is stored in the file

The C2PA specification stores the record as JUMBF data and puts it in a place that depends on the file type:

• JPEG: in APP11 segments.
• PNG: in a chunk named caBX.
• WebP: in a chunk named C2PA. WipePic looks for this chunk.

Because the record is stored next to the picture and not inside it, a tool can remove it without touching the pixels.

What WipePic removes and what it keeps

WipePic removes the embedded C2PA manifest and, in the same step, the other hidden data in the file: location, camera and phone details, dates, author fields, editing history, AI labels, PNG prompts and settings, text notes and thumbnails.

It keeps what the picture needs to look right: orientation, color profile, and, where the file has them, transparency, animation and an HDR gain map. It does not re-encode the image, so the pixels do not change. After cleaning, WipePic reads the new file again and does not save it if a C2PA record or other removable metadata is still in it.

What can still identify the image

• Watermarks in the pixels, such as SynthID. They are part of the image. See does removing metadata remove SynthID.
• Records stored outside the file. Adobe says that Content Credentials stored in the Adobe Content Credentials cloud can be recovered even from a screenshot (Adobe Help Center, updated September 2025). That is a lookup on Adobe’s side. WipePic cannot remove it, and a match is not certain for every image.
• Copies that you or others posted before. They keep their own data.
• Other signals on a platform. A site can use its own checks to label an image.

The LinkedIn Content Credentials icon

LinkedIn Help says that it shows the C2PA icon on image and video content that is cryptographically signed with Content Credentials. This is the small “CR” icon. If you post a file that carries C2PA credentials, the icon is visible to everyone who sees your post. A viewer can click it to see the app or device used, who issued the credential and, where it applies, whether AI was used. The icon alone does not say that an image is AI-made. LinkedIn also says that it cannot yet label all AI-generated content.

If the manifest is not in the file that you upload, LinkedIn has no record in the file to show. Rules about telling people that an image is AI-made still apply to you. WipePic does not change them.

How to remove Content Credentials

1. Add your images to the tool above. You can also add a folder.
2. Look at the list of what WipePic found. Content Credentials (C2PA) appear by name.
3. Click Remove hidden data and download the cleaned copies. Several files come as one ZIP.

To check a file first, use the AI metadata checker. For a wider clean-up of AI images, see remove AI metadata.