Numbers Loading Icon

Combating AI Misinformation: Capture's Integration of C2PA Provenance Data

Posted in:

Combating AI Misinformation: Capture's Integration of C2PA Provenance Data

In today's digital landscape, the line between reality and deception has blurred, leaving the general public grappling with the increasing challenge of distinguishing truth from falsehood. The pervasive spread of "fake news" and the rise of sophisticated AI tools, capable of generating convincing deep fake videos and manipulated images, have fueled a climate of skepticism. Daily, we bear witness to alarming instances, AI generated images showing fake explosion outside the Pentagon, manipulated images projecting false statements from foreign leaders and many others. The result is a disconcerting atmosphere where trust in what we see, hear, or read online has eroded.

This era of uncertainty prompts a critical question: How can individuals navigate through the noise of misinformation and establish a clear distinction between genuine and manipulated media? Recently, tech industry giant, Microsoft joined the fray by announcing the latest AI improvement to their product, it’ll add C2PA standard Provenance Data to any AI generated images. Before this big move from Microsoft, Capture has already made it possible by integrating C2PA Provenance Data to every picture uploaded to the platform. 

Introduction to C2PA Provenance Data

What is C2PA, and Who Adopts It?

The Coalition for Content Provenance and Authenticity (C2PA) is a collaborative effort by industry leaders, including Microsoft, Adobe, Intel, and Sony that is working to develop standards for verifying the authenticity of digital content. One of the key technologies that C2PA is promoting is Provenance Data.

Provenance Data are invisible markers that can be embedded in digital content. These markers can be used to identify the creator of the content, track its provenance, and verify its authenticity.

Benefits of C2PA Provenance Data

The Coalition for Content Provenance and Authenticity (C2PA) Provenance Data offers several benefits, contributing to enhanced transparency and authenticity in digital media. Here are some key advantages:

  1. Authentication and Validation:
  • The C2PA Provenance Data provides a reliable method for authenticating and validating digital files. This is crucial in an era where deep fake videos, manipulated images, and deceptive media are prevalent.
  1. Tamper-Evident Records:
  • C2PA utilizes public key infrastructure (PKI) to create tamper-evident records for digital content. This means that any attempts to alter or manipulate the content can be readily detected, ensuring the integrity of the media.
  1. Trust and Confidence:
  • By implementing C2PA Provenance Data, content creators, distributors, and consumers can have increased trust and confidence in the authenticity of the media they encounter. This is particularly important in combating misinformation and enhancing overall content provenance.

Case Studies: Real-World Scenarios of C2PA in Action

Media and Entertainment Industry

  • Authenticity in Digital Assets : Film studios can use C2PA to authenticate and validate movie trailers, posters, and promotional images, ensuring audiences receive genuine content without falling victim to manipulated marketing materials.
  • Protection Against Deep Fakes : News outlets can employ C2PA to verify the authenticity of video clips, safeguarding against the spread of deep fake news and maintaining the integrity of journalistic content.

E-commerce and Retail

  • Trust in Product Imagery : E-commerce platforms can utilize C2PA to having Provenance Data in the product images, providing customers with confidence in the authenticity of items showcased online and reducing the risk of deceptive marketing.

Legal and Law Enforcement

  • Tamper-Proof Evidence : Law enforcement agencies can use C2PA to create tamper-evident records for digital evidence, ensuring its integrity and authenticity in legal proceedings.

Publishing and Journalism

  • Verified News Articles : Publishers and news organizations can employ C2PA to validate the authenticity of news articles, photos, and videos, fostering trust among readers in the midst of a "fake news" era.

Capture's Role in C2PA Integration

Create C2PA With No Technical Jargon 

Capture introduces simple-to-use, scalable C2PA provenance tools across its product suite, including Capture Cam, Capture Dashboard, Capture API and Capture Eye, making them ideal for businesses and individuals of all sizes.

Capture API is a powerful software development kit that allows developers to easily integrate Capture's C2PA provenance tools into their own applications. Capture Dashboard is a web-based platform that allows users to easily upload, manage, and track the provenance of their digital assets with C2PA Provenance Data. Capture Cam is a mobile app that allows users to add C2PA Provenance Data to images and videos, making them ideal for businesses and individuals of all sizes.

Every Uploaded Image Has C2PA Injection

Every image that is uploaded to Capture Products is automatically injected with a C2PA Provenance Data. This means that users can be confident that their images are authentic and that they can track their provenance.

Immutable Provenance on the Blockchain

Blockchain technology is very ideal for provenance because it is immutable. Once digital assets are recorded on the blockchain, their ownership history is permanently and publicly recorded.  Capture integrates C2PA Provenance Data with the blockchain to give users and businesses indisputable evidence of their digital assets' lineage, building trust and credibility in the digital age. 

Demo: How to Add C2PA Provenance Data to Images Using Capture

Let’s try adding your images to Capture! Head over to our dashboard ( Start by creating a free account if you haven’t already yet. Once in, just click on create, upload the images and fill out some details of your choice.

Once done, head over to ‘Manage’ and you can click on your registered picture. In this case, you can check the details of the picture we uploaded above here.

This can be done with AI generated images too, immediately after their creation if you wish. We have also made a tutorial. (How to Create Traceable Generative AI in an Era of Untraceability)

The content you registered with Capture is intact and has no content metadata injection at the time of registration. In order to see the content with C2PA Provenance Data, you can download it via Capture API. (Download C2PA Asset)

Now that we have downloaded the C2PA signed asset, we can verify the content credential with the CAI (Content Authenticity Initiative) site (

We can see the image we uploaded here has not yet been altered in any way. Now let’s try to do a simple editing on the image.

The image has been added with some random scribble with an editing tool, let’s re-check the new, altered image at C2PA Verify site.

In the Content Ingredients window on the left, you can see the original images, along with the alteration we did previously. This allows you to see the history starting from the birth of this image, to the several changes that have been made to it.

In the near future, new features will be added to Capture Dashboard and Capture Cam, allowing users the option to add C2PA Provenance Data directly without API.

Implementation & Integration

Capture Products' tools are easy to integrate into existing workflows. Businesses and individuals can use the Capture API or services such as Capture Cam and Dashboard to add C2PA Provenance Data to their images and videos. To learn more about Capture API and Capture Services, feel free to reference related materials here

Decentralized and Immutable: Why Do We Need Decentralized Records Besides C2PA?

While C2PA Provenance Data are widely accepted as a digital provenance standard, its mutable nature and the challenges with verifying records calls for a more immutable and decentralized solution. Capture Products do not limit themselves to C2PA Provenance Data and extend their services to include blockchain provenance records. 

Every image and video registered with Capture is indexed with a unique id (Nid) and has its provenance records recorded on-chain. Since these records are on the blockchain, they are both immutable and decentralized, meaning it cannot be alternated or deleted and most importantly it can be easily verified. Blockchain records provide an additional layer of security and transparency for users of Capture Products.

The Ethereum Improvement Proposal (EIP) to add C2PA consent data, such as the Content Consent for AI/ML Data Mining, to on-chain metadata is under review. Comment to help create a better standard.

Roadmap & Future Enhancements

In the coming months, Capture will be expanding its digital management capabilities as well as improving its overall performance and stability in the hopes of elevating users' content authenticity experience. Capture’s development is far from complete as we strive to provide the most comprehensive digital provenance solutions. Stay tuned! 

Beyond Just Provenance - Authenticity in the Age of AI 

Generative AI represents a remarkable leap forward in creative technological capabilities. We can’t stop technologies from progressing, not that we even think about it. But it's absolutely crucial for everyone to get the hang of spotting the real deal from the fakes. At the same time, content creators need top-notch tools to back up their genuine content. The collaboration between Numbers Protocol and Instill AI redefine the data landscapes and unlock AI & blockchain synergies. Watch the full webinar now on YouTube and deep dive into the future of digital transparency.


Although Capture, just like C2PA, is still in its early stages, as it gains more adoption and traction, this standard is poised to become a cornerstone in establishing digital trust in our rapidly evolving AI-centric world. We invite you to join us on this transformative journey, where authenticity and provenance redefine the future of visual content.

You're signed up! Watch you inbox for updates.
Oops! Something went wrong while submitting the form.