Volume 16, Number 6
Challenging Multi-Vector Exclusion by Improving ROI Adoption in Applied AI Systems: The E.A.S.E. Framework as a Systemic Evaluation Tool for Intersecting AI Bias and Digital Exclusion
Authors
Kevin J. Spellman, Tax & Legal - Digital Innovation, UK
Abstract
This paper addresses the core challenge of multi-vector exclusion in applied AI systems, arguing that the industry's singular focus on Return on Investment (ROI) constitutes a systemic failure in evaluation (Chen, 2025). This crisis is exacerbated by the progression to autonomous agentic systems, which intensify the risk to human agency by relying on workflows that have not been adequately understood. Mixed-methods research reveals distinct, intersecting harms across two axes: Neurodiversity/Disability (evidenced by systemic technical friction and the "disability tax") and Racial Bias(evidenced by significant algorithmic preference bias and job displacement risk for Black heritage professionals). These findings demonstrate that AI adoption is separating functional rollouts from the critical need for co-creative change management. To resolve this, the paper introduces the E.A.S.E. Framework (Equity of Access, Agency & Participation, Succession Continuity, and Employment Impact)—a novel governance methodology designed to integrate human-centered, auditable metrics across the product lifecycle, ensuring technology supports equitable and resilient organizational transformation.
Keywords
Multi-vector Exclusion, Algorithmic Bias, AI Governance, Digital Exclusion, Agentic Systems