A Multi-Layered Research Framework for Human-Centered AI: Defining the Path to Explainability and Trust

AI transparency requires a multi-layered approach: The paper demonstrates that effective explainability isn't achieved through a single mechanism, but rather through combining built-in model transparency, customized explanations for different users, and ongoing feedback loops that improve explanations over time.

One-size-fits-all explanations don't work: The framework's second layer tailors explanations based on individual cognitive load and expertise level, suggesting that a doctor, financial advisor, and software developer need different types of explanations for the same AI decision—a finding relevant to making AI trustworthy across diverse professional contexts.

Real-time feedback improves AI trustworthiness: The dynamic feedback loop component allows AI systems to learn from user interactions and refine their explanations in real-time, indicating that trust in AI systems can be actively built and strengthened through continuous dialogue rather than static documentation.

Explainability is critical for high-stakes decisions: Testing across healthcare, finance, and autonomous systems shows that transparency and interpretability directly impact regulatory compliance and public trust, suggesting that organizations deploying AI in consequential domains cannot treat explainability as optional.

Human-centered design and explainability must work together: The paper argues that aligning AI with human values (HCAI) and making AI decisions understandable (XAI) are complementary goals that strengthen each other, rather than separate concerns—a unified approach the authors call Human-Centered Explainable AI (HCXAI).