About Helix Fabric

Helix Fabric is a distributed verification infrastructure purpose-built to detect synthetic identities — AI-generated organizations that pass legacy due diligence checks.

Invented and Published by Thomas Perry Jr.

Thomas Perry Jr. is a security researcher specializing in synthetic identity detection and verification infrastructure. His published methodology forms the foundation of every detection algorithm in Helix Fabric, with 8 peer-reviewed publications deposited on Zenodo with persistent DOIs.

Technology Overview

The Problem

Generative AI has made it trivial to create convincing organizational facades — complete with websites, social media presence, executive profiles, and press coverage. These synthetic entities pass traditional KYC/KYB due diligence because legacy verification systems check for the presence of identity artifacts, not their authenticity. Organizations making trust decisions based on surface-level checks are exposed to fraud, counterparty risk, and reputational damage from engagement with fabricated entities.

The Solution

Helix Fabric is a distributed verification infrastructure that goes beyond surface-level checks. It analyzes the structural coherence of an entity's web presence, examines content provenance chains for authenticity gaps, maps entity relationship graphs to discover closed reference loops, and scores trust across five weighted dimensions. The result: actionable verdicts that distinguish genuine organizations from AI-generated facades, delivered in under 200 milliseconds.

Architecture

Queue-driven, event-sourced architecture with distributed scanning and real-time trust scoring.

1

Queue-Driven Scanning

Targets are submitted via API and enter a queue-driven pipeline. The sweep process selects the stalest targets with a configurable freshness guard, dispatching scan_requested events to the queue consumer.

2

Multi-Signal Detection

24 distributed workers execute parallel scans across three signal categories: synthetic web presence (7 types), C2PA provenance gaps (6 types), and ecosystem/Möbius loops (4 types). Each signal carries a confidence score and evidence chain.

3

Trust Scoring

A 5-dimension weighted scoring model (entity, compliance, behavioral, counterparty, ethics) produces a composite trust score multiplied by an ethics factor. Verdicts are classified as allow, require_approval, require_review, or refuse.

4

Continuous Monitoring

Real-time WebSocket feeds broadcast scan completions to connected dashboards. Automated re-scans maintain ongoing surveillance of the entire entity graph, with auto-discovery recommending new targets based on relationship analysis.

Enterprise Services via SignaBuilder

Enterprise deployments, custom scanner development, SLA-backed service agreements, and white-label options are available through SignaBuilder. SignaBuilder provides the operational wrapper for Helix Fabric deployments at organizational scale, including dedicated account management, custom integration support, and compliance documentation.

signabuilder.com

Research Foundation

Every detection algorithm in Helix Fabric is grounded in published research with persistent identifiers. The methodology is fully transparent and citable.

All publications are open access and archived on Zenodo with persistent DOIs. Author identity verified via ORCID 0009-0007-1476-1213.

Ready to Start?

Contact us to discuss your entity verification needs or request API access.