Enabling ethical AI research through provenance disclosure

Truepic worked with researchers from the Affective Computing group at MIT Media Lab to add secure, cryptographic provenance to over 30 deepfake videos.

View the Research Project Here

Secure disclosure for ethical AI research

Safe, scalable research is critical to understanding the impacts of generative AI and synthetic media. As research on generative AI and synthetic media grows, content transparency can help researchers inform participants and add lasting attribution to synthetic content. This kind of cryptography-powered disclosure and attribution helps to reduce the risks of synthetic media being taken out of context after it is used in academic research.

Sign each image and video

Truepic worked with researchers from the Affective Computing group at MIT Media Lab to add secure, cryptographic provenance to over 30 deepfake videos.

Display provenance details

The research team used provenance to debrief participants in their study and ensure these synthetic media files were traceable back to their institution.

Transparency for a growing field of study and innovation

Using the C2PA open standard, the researchers were able to disclose and attribute the AI-generated videos they had shown to participants during the experiment. Correct attribution, sealed into each video file, helps reduce the risk that videos used in an experiment will be taken out of context.

Learn more about the C2PA

Content TransparencyA first-of-its kind example. A growing best practice.

Informing participants is crucial to ethical research, especially research that involves hyperrealistic, often deceptive synthetic content.

Disclosure

This first-of-its kind approach pioneered by Truepic and researchers at MIT Media Lab means that viewers do not have to guess if what they are looking at was AI-generated.

Learn why disclosure matters in deepfake research

Attribution

With transparent, tamper-evident content provenance data, research institutions can ensure that synthetic content is correctly attributed and traceable, back to them.

Check out the study’s initial findings

Interested in learning more? We’d love to hear from you.

Fill out the form below to subscribe.

Request more information