Technologies such as generative adversarial networks (GANs) and other machine learning techniques have made it easy for malicious parties to create forged or fake content such as images and videos. It is difficult to determine the provenance of such content. This disclosure describes techniques that utilize a notarization system based on a blockchain ledger. New content is added to the system along with a secure time code that marks the first occurrence of the content and hashed content identifiers. The use of timecode imposes a time limitation within which malicious actors need to generate forged or fake content, which along with the high computational or human cost of creating forged content places limits on the quality of forged content. Such forged content is detected based on associated time codes and by identifying inconsistencies or errors in the content.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.