Google has announced steps to increase transparency for AI-generated content by implementing the latest Content Credentials technology developed by the Coalition for Content Provenance and Authenticity (C2PA).
Google, which joined the C2PA as a steering committee member earlier this year, has been instrumental in developing version 2.1 of the Content Credentials technical standard. This latest version boasts enhanced security against tampering attacks, ensuring the integrity of provenance information attached to digital content.
The tech giant plans to integrate this technology into several key products over the coming months. In Google Search, the "About this image" feature will display C2PA metadata for images, indicating if they were created or edited with AI tools. Google's ad systems will begin incorporating C2PA metadata to inform policy enforcement. Additionally, YouTube is exploring ways to relay C2PA information for content captured with cameras.
These implementations will utilise the forthcoming C2PA Trust list to validate content origin, adding an extra layer of authenticity to the provenance data. By incorporating this technology across its platforms, Google aims to provide users with a comprehensive understanding of the content they encounter online.
This initiative aligns with ongoing projects like SynthID, an embedded watermarking technology created by Google DeepMind, and the company's participation in various AI safety coalitions and research groups.