Research
Identra Research investigates the hard problems at the intersection of identity, fraud, and privacy. We develop and share new approaches that make online identity verification more accurate, fair, and safe for everyone.
Core research areas
Fraud intelligence
Analyzing, developing, and improving fraud detection and prevention methods globally by discovering new methodologies.
Online safety
Exploring how to create safer online environments, including cryptography and privacy-preserving age verification.
Accessible identity
Designing accessible and inclusive authentication and identity verification systems that expand access to digital services for all.
Data privacy
Furthering the goal of protecting users' data during identity verification, fraud prevention, and age assurance.
Publications
Scientific papers, white papers, and technical briefings authored by Identra engineers and research partners.
Layout-Aware Representation Learning for Open-Set ID Fraud Discovery
We develop a layout-aware representation learning framework for identity documents that maps each document into a discriminative 512-dimensional embedding space with strong inter-class separability and intra-class compactness.
Authors
Identra Research
From Measurement to Mitigation: Quantifying and Reducing Identity Leakage in Image Representation Encoders with Linear Subspace Removal
We present a comprehensive study of identity leakage in visual embeddings and introduce Identity Sanitization Projection (ISP) as an effective mitigation. This work takes a step toward making powerful vision models more privacy-friendly, which is crucial for real-world deployment.
Authors
Identra Privacy Lab
Identity Muling and Techniques to Combat the Rise of Second-Party Fraud
As identity verification systems improve at detecting deepfakes and synthetic identities, fraudsters are increasingly turning to identity muling, a form of second-party fraud that relies on real people willingly sharing their identities or completing verification on behalf of bad actors.
Authors
Identra Fraud Intelligence
Threat Domain Partitioning and Sorted Rejection Labeling: Benchmarking for Adversarial Environments
Structural differences in fraud detection make conventional classification metrics both hard to compute and often misleading. We introduce a practical framework that defines business-aligned benchmarks that are cheaper to label, stable under performative drift, and foundational for systematic and iterative fraud model development.
Authors
Identra Research
Modernizing Know Your Business (KYB) with online credibility signals
Traditional Know Your Business (KYB) checks confirm a company's legal existence but often miss signs of real-world legitimacy or fraud. This white paper explores how integrating online credibility signals from a business's website, social media presence, and customer reviews adds vital context to KYB.
Authors
Identra Business Identity Lab
Understanding and Verifying Digital Identity
Digital identity connects who people are in the real world with how they interact online, combining personal data, behavioral patterns, and technological signals to establish authenticity. As digital ecosystems expand and AI-driven threats evolve, strong identity verification is essential.
Authors
Identra Identity Lab
Navigating Global Age Verification Laws in the Digital Era
As regulatory bodies worldwide strengthen age verification and data privacy regulations, organizations must balance compliance with seamless user experiences. Understanding emerging laws and implementing adaptable, privacy-conscious verification systems are key.
Authors
Identra Trust Research
Reading list
A curated database of global regulatory trackers, industry briefings, and advanced research to help compliance and risk engineering teams stay ahead of the curve.
Atlas: A Global Age Regulation Tracker
Identra Atlas is a free, collaborative database and API designed to help compliance teams navigate the complex landscape of age assurance and privacy laws.
Curated Briefings & Reports
Explore, bookmark, or complete high-impact analyses of biometric safety and dynamic trust landscapes.
The Age Verification Revolution: Balancing Child Safety and Data Privacy
An in-depth exploration of state-of-the-art cryptographic age assurance techniques, exploring zero-knowledge proof protocols and double-blind validator models.
Understanding Biometric Injection Attacks in Modern Face Verification
Visualizing the shift from static presentation attacks (masks, printed photos) to hardware/software virtual cameras injecting synthetic video loops directly into the browser.
EU Artificial Intelligence Act: Implications for Biometric Systems
A clear, structured compliance roadmap analyzing which automated verification pipelines fall under 'High Risk' or 'Prohibited' categories of the new EU AI Act.
Biometrics in the Wild: Evaluating Real-World Facial Recognition Bias
A collaborative demographic audit assessing False Match Rates (FMR) and False Non-Match Rates (FNMR) across global cohorts to refine algorithmic fairness.