Login.gov: artificial intelligence technology used for detecting and mitigating fraud for remote unsupervised identity verification
Description
The AI is intended to solve the problem of detecting and mitigating fraud, spoofing, or deepfakes introduced by the remote identity proofing process. Specifically, it aims to compare and verify a user's identity by scanning identity documents uploaded online for authenticity, conducting one-to-one facial matching comparing the applicant’s self photograph (selfie) with the image on the identity evidence, performing liveness detection to mitigate imposters uploading images of stolen or counterfeit documents, and fraud detection analytics, all with the goal of ensuring the applicant is the person they are claiming to be. This is also to achieve NIST SP 800-63 compliance, which effectively requires AI-based capabilities for compliance with remote proofing.
AI / analytics pattern
Computer Vision
Automation level / stage
c) Deployed - The use case is being actively authorized or utilized to support the functions or mission of an agency.
Expected benefit
Efficiency, productivity, service quality
Controls / human review
Inventory row should be validated against agency governance and M-25-21 controls.