Solutions to Deepfakes: Can Camera Hardware, Cryptography, and Deep Learning Verify Real Images?
Alexander Vilesov, Yuan Tian, Nader Sehatbakhsh, Achuta Kadambi
TL;DR
The paper addresses the looming challenge of authenticating camera-captured images in the era of highly realistic synthetic media. It compares detection-based verification with cryptography-based provenance, arguing that cryptographic methods like PKI/C2PA offer more durable solutions against advancing generative models. It details practical paths forward, including camera signing workflows, adoption steps, spoofing countermeasures, and the creation of new file formats to simplify searching for real images. The work highlights the necessity of co-design across cryptography, hardware, and machine learning to safeguard digital media and outlines concrete actions to move toward verifiable authenticity.
Abstract
The exponential progress in generative AI poses serious implications for the credibility of all real images and videos. There will exist a point in the future where 1) digital content produced by generative AI will be indistinguishable from those created by cameras, 2) high-quality generative algorithms will be accessible to anyone, and 3) the ratio of all synthetic to real images will be large. It is imperative to establish methods that can separate real data from synthetic data with high confidence. We define real images as those that were produced by the camera hardware, capturing a real-world scene. Any synthetic generation of an image or alteration of a real image through generative AI or computer graphics techniques is labeled as a synthetic image. To this end, this document aims to: present known strategies in detection and cryptography that can be employed to verify which images are real, weight the strengths and weaknesses of these strategies, and suggest additional improvements to alleviate shortcomings.