VORNAC RESEARCH
Machine-learning security, blockchain, and data-layer threats.
Background
Vendor-neutral landscape map: model families, training pipelines, deployment patterns — plus which statistical/ML models fit which security-analytics problems and where they reliably fail.
Natural-language processing applied to security work: log clustering, phishing detection, report summarization, and where modern LLM-driven techniques fit (and don't).
Smart-contract, bridge, and consensus-layer threat classes — where the field's actual losses cluster, and the audit patterns that catch them.
Practitioner-level hashing reference: when collision resistance matters, when length-extension bites, and what to pick today.
From reference to evidence