Abstract

Artificial Intelligence (AI) has progressed from rule-based systems to sophisticated machine learning models that can automate
complex tasks. While these advancements have reshaped industries, the growing dependence on algorithmic decision-making
raises concerns about fairness, transparency, and alignment with human values. This paper argues for a paradigm shift toward
Human-Centered AI (HCAI), an approach that prioritizes human well-being, ethical responsibility, and collaborative intelligence
over pure automation. We examine the shortcomings of automation-driven AI, including issues of bias, lack of interpretability,
and reduced human agency. By integrating principles of explainable AI, participatory design, and value-sensitive computing,
HCAI ensures that intelligent systems remain trustworthy, inclusive, and aligned with societal goals. Through case studies in
healthcare, education, and governance, we demonstrate how Human-Centered AI fosters accountability, improves decision
quality, and strengthens human-machine collaboration. Finally, we propose a research roadmap for embedding human values
into AI design, moving beyond algorithmic efficiency toward responsible and sustainable innovation.