Inference Authority
A governance vocabulary for authoritative and non-authoritative inference pathways, divergence validation, and controlled authority transitions.
READ THE FRAMEWORK04 / Research & IPNot a black box.
Not a black box.
A body of work.
Tavri develops the language, open artifacts, and patent-pending architectures required to make AI authority measurable, reviewable, and deployable.
Selected publicationsMake control
Make control
measurable.
Technical work for engineers, operators, and regulated environments.
FAC-008 Asset Intelligence
How AI-assisted nameplate extraction can support traceable facility-rating inputs while responsibility remains with the asset owner.
READ THE PAPERTavri Asset Intelligence Schema
An implementation-agnostic JSON Schema for structuring asset evidence, provenance, and recorded engineering artifacts.
VIEW THE OPEN SPECInference Authority Patent Family
Published and pending applications covering governance of AI authority transitions, failover, validation, and secure field intelligence.
VIEW THE PATENTSStandards & policyContribute to
Contribute to
the control language.
Tavri contributes to emerging standards and federal discussions on governance of AI systems, including authoritative output control and model-transition safety.
In March 2026, founder Joshua A. Wright submitted recommendations to NIST on authority gating, non-authoritative parallel evaluation, controlled transfer, revocation, deterministic enforcement, and auditability.
VIEW NIST SUBMISSION