The Content Authenticity Initiative (CAI) is a cross-industry community of major media and technology companies, civil society, and many others. It was founded by Adobe in 2019, and Adobe continues to lead it. The CAI develops open-source tools for verifiably recording the provenance of any digital media, including content made with generative AI. The community's mission is to support broad adoption, making content authenticity and transparency scalable and accessible.

The Coalition for Content Provenance and Authenticity (C2PA) is a formal coalition dedicated to addressing the prevalence of misleading information online through the development of technical standards for certifying the source and history (or provenance) of media content. It’s a mutually governed standards development organization (SDO) under the structure of the Linux Foundation’s Joint Development Foundation, formed through an alliance between Adobe, Arm, Intel, Microsoft, and Truepic.

The C2PA unifies the efforts of the Adobe-led CAI, which focuses on systems to provide context and history for digital media, and Project Origin, a Microsoft and BBC-led initiative that tackles disinformation in the digital news ecosystem.

The CAI drives feature explorations, creates open-source tools, and fosters community around a movement toward greater content authenticity online. The C2PA complements these efforts by providing the end-to-end open technical standards for creators, editors, publishers, media platforms, and consumers.

The CAI has built a cross-industry community with diverse viewpoints to foster discourse and action around bringing content authenticity into a global practice. Collaborators include Adobe, the Associated Press, BBC, Microsoft, The New York Times Co., Reuters, Leica, Nikon, Canon, Pixelstream, Truepic, Qualcomm, and many others, including members of civil society.

Content Credentials are the verifiable metadata that the CAI’s tools help implementers generate, read, and display, based on the C2PA’s technical standards. They are essentially a nutrition label for digital content, and can contain information about who produced a piece of content, when they produced it, and which tools and editing processes they used.

The information embedded in Content Credentials depends on a combination of the information that each implementer chooses to support and the information that the person applying the Content Credentials chooses to include. The manifest that defines the Content Credentials can include various assertions about the associated content, including the other content ingredients used to produce it; the date, time, and location where it was produced; and the device or software that was used in the production process.

Content Credentials and the CAI’s tools implement the open standards created by the C2PA. The CAI’s tools for Content Credential generation, display, and inspection rely on proven cryptographic methods for protecting information from being tampered with, in alignment with C2PA standards. We also have an open membership model that encourages feedback from members. Learn more about Content Credentials.

Unlike other types of metadata, Content Credentials are cryptographically signed, making them tamper-evident. In other words, if someone makes changes to the associated content or the attached data, you’ll be able to tell that alterations were made after the content was signed.

While solutions like Content Credentials are sometimes confusingly described as “watermarks”, watermarking has a specific meaning that’s only one piece in the multi-pronged approach to resilience represented by Content Credentials. 

Content Credentials may combine watermarking, secure metadata, and digital fingerprinting to offer a comprehensive solution for labeling digital content. Learn more.

Content Credentials provide information about the origin and history of a piece of content, but they aren’t intended to prescriptively indicate whether a piece of content is “real.” For example, if an AI tool supports Content Credentials, then Content Credentials can indicate that an image was generated with AI. If an image was taken with a C2PA-enabled camera, the resulting Content Credentials would show that. And if any subsequent edits were made with C2PA-enabled software tools, that information could also be captured.
 

Content Credentials serve as a kind of nutrition label for digital content. When available, they can help you be more informed about the content you’re consuming, allowing you to form your own opinions and conclusions about it.

There are generally three ways to address misinformation: education, detection, and attribution. The CAI is creating an attribution- and transparency-based solution by supporting the generation, display, and inspection of Content Credentials with our tools. Content Credentials can help capture and convey to consumers information about who was involved in the production of a piece of content and how it was produced. 

We believe that media literacy is an additional key component of addressing mis- and disinformation, and we have worked with education experts to prepare robust curricular materials. The CAI undertakes efforts around education where we’re able, and we’ll be transparent about information that can aid in detection. Ultimately, however, addressing the complete problem of misinformation requires ongoing collaboration across the digital content ecosystem.

The CAI doesn’t enforce permissions for access to content. The name displayed in Content Credentials generated with the CAI’s tools might not always be the same as the individual or organization that owns the content. Additionally, creators aren’t required to include an identity in their Content Credentials, and when they do, that identity might not always be displayed depending on the publishing or viewing platform.

While Content Credentials are compatible with blockchain, they don’t require or use blockchain directly.

CAI community members have implemented solutions that incorporate blockchain. Here are few examples:

• Starling Lab uses blockchain for distributed storage.
• Numbers Protocol uses blockchain to store Content Credentials — see Numbers Blockchain.
• The Click mobile app integrates Content Credentials and leverages the Ethereum Blockchain designed to store and verify the origin of images created with the app. 

Content Credentials implement the C2PA’s technical standard and cryptographic methods for protecting information from being tampered with, making them tamper-evident. When information in Content Credentials has been improperly changed, or when the associated content has been changed after Content Credentials were applied, the Content Credentials will identify and display this information. This system protects against the fabrication of a provenance chain, and will make evident any efforts to do so. Inspection tools like the Verify website will show this information so viewers can determine for themselves whether they can trust the content.

Protecting privacy is one of our guiding principles. We recommend that users apply Content Credentials on an opt-in basis, and we want to emphasize that including identity information like the creator’s name should always be optional. Users can also control the other categories of data they want to include in their Content Credentials, such as the editing actions and other content ingredients they used in their production process. For those with unique privacy needs (e.g., human rights workers and journalists), Content Credentials support the ability to redact sensitive information later on.