How To Measure Deepfake Detection

Deepfakes keep getting more sophisticated and harder to catch. How do you know if your detection systems can keep up with the pace of this AI-powered fraud?

Now there is a way to test it. The U.S. Department of Homeland Security funded research that created a dataset called IDNet, which consists of more than 800,000 AI-generated synthetic documents. The goal of this initiative was to benchmark document verification solutions by seeing how well they can detect these AI-generated deepfakes. 

We tested Microblink’s technology against IDNet, and so can you. Download this report to learn:

• What the IDNet dataset reveals about the future of AI-generated fake documents
  
• How the availability of such a dataset can transform deepfake detection
  
• How Microblink’s detection performed under extreme fraud scenarios