OpenAI has announced that it will add watermark to the metadata of the artificial intelligence (AI)-generated images created by DALL-E 3. The company stated that it will now use the open technical standard adopted by the Coalition for Content Provenance and Authenticity (C2PA), and add the information that the image was generated by AI, the name of the AI tool, and the name of the app used to create the image in its metadata. The move comes as Meta announced that AI firms need to adopt a common standard to help detection tools identify and label AI content on its social media platforms.
In a post, OpenAI revealed the move, as well as technical details around it. It said that images generated with ChatGPT on the Web client and API, which uses the DALL-E 3 model, will now contain a new metadata as per the C2PA standard. The same watermark process will be rolled out to the ChatGPT app by February 12. C2PA standard is a specific watermarking technology which adds a stamp on the image itself and embeds the information inside the image as well. As a result, a CR symbol can be seen on the top left of the image and a detailed version can be checked in its metadata.
Through the metadata, users can check the origins of the image, including information on the AI model and the app used to create it. In the examples shared by OpenAI, the metadata shows a content summary which says, “This image was created with an AI tool.” A separate tab for Process shows whether an API, Web client, or ChatGPT was used, in addition to showing the underlying AI model. As per the company, adding the metadata may slightly increase the size of the image, but there will be no effect on the quality.
While this makes the image more secure than just a visual marker, there are still ways to bypass it. OpenAI highlighted that many social media platforms remove the metadata from uploaded image, and taking a screenshot of the image will also remove it. Therefore, this method may not be enough to determine if an image was indeed created by DALL-E 3 or other AI models.
C2PA includes companies such as Adobe, Microsoft, BBC, Sony, Leica, Nikon, and more. It has been pushing for the adoption of this technology as a method to detect and correctly label AI-generated content. The CR symbol, created by Adobe, was also given by the same group.