Evidence-Bound Adversarial Verification (EBAV): A Governed Field Method for AI-Assisted Software Verification and Release
Year: 2026
Type: article
Abstract: EBAV v0.1.0 is a proposed seven-stage field method for evidence-bounded claims, immutable candidates, independent adversarial software review, regression-driven correction and separately scoped human authorization. Includes a retrospective Jesus Is King case account with explicit evidence limitations. Not peer reviewed or certified.
Author: Coleman, Brendon R.
Language: English
Subfield: Safety, Risk, Reliability and Quality
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7233175941
