Google is building an AI image detection feature into Messages that would let users check whether photos shared in a conversation were created or edited using AI, according to Android Authority. The feature is not live yet, but code strings found in a recent beta release of the app point to a C2PA (Coalition for Content Provenance and Authenticity) Content Credentials integration.
More than a simple AI label
C2PA is an industry-standard system for logging how media is created and modified. Content Credentials work by embedding metadata directly into an image file, recording details such as the creator, the software used, and any edits made along the way. Baking that into Google Messages would give users a straightforward way to assess the origin of images they receive. What stands out about the implementation is that it won’t just stick to a binary AI or not-AI verdict.
Code strings highlight labels such as “Media made with AI,” “Edited with multiple AI tools,” “Parts of this media were made with AI,” and “Media captured with a camera without software adjustments,” among others. That level of detail would make this feature more useful than a blanket warning label.
Still in development
Google has already built SynthID watermarking into Gemini to help users identify AI-generated content across Android. Unlike SynthID, which is Google’s own proprietary tool, C2PA is an open industry standard backed by several companies including Adobe, Microsoft, and Qualcomm, making it better suited to flag AI content from any source.
Android Authority says the feature will likely surface through a “View details” option in the overflow menu that appears after tapping on a shared image in a chat. However, details about the feature come from an APK teardown, meaning it is not functional yet and may change before any public release. Google has not made anything official yet, so it’s difficult to say when it will roll out to users.
