The Pedigree of Provenance: Patented Emotion AI
Where Empirical Science Meets Enterprise Infrastructure.
This repository contains the foundational research, technical briefs, and primary patent filings governing VERN OS—the industry’s only deterministic, single-shot emotion analytics engine.
Unlike modern trend-chasing API layers that rely on expensive, high-latency multi-agent loops, VERN AI was forged through a decade-long engineering evolution. In 2016, operating as THiNC. Technology, our team engineered the stateful web framework used by Dr. Saleem Alhabash’s digital media labs at Michigan State University to study human-bot interaction boundaries and deception.
It was within this live-fire testing crucible that our team isolated the “Emotion Gap”—discovering that conversational automation fails human trust thresholds not from a lack of data, but from structural emotional blindness.
Available Documentation:
The VERN OS Technical Briefing: An executive-ready deep dive into our architectural lineage, detailing how we completely eliminate the enterprise “token tax” while maintaining emotionally aligned, deterministic AI behavior.
US Utility Patents (No. 11,138,387 & 11,791,995): The core intellectual property specifications authored by Craig Tucker and Bryan Novak. These filings document our proprietary mathematical formulas for calculating real-time Incongruity Scores, Linguistic Prediction Errors, and Emotional Velocity.
Associated Peer-Reviewed Literature & Datasets: Academic references validating our decade of work in computer-mediated communication (CMC) competence, psychophysiological response tracking, and affective subtext.
Explore the documentation below to see the mathematical and empirical framework that diagnosed the industry’s architectural flaws a decade ago.