A Universal Meta-AI Verification Framework (UMVF) Based on Cross-Sensor Correlation and Reliability Analysis

Ensuring the robustness and reliability of complex Artificial Intelligence (AI) systems presents a critical challenge. This work introduces a hierarchical universal Meta-AI Verification Framework (UMVF) to address limitations inherent in single-sensor validation methods. The UMVF acts as a supervisory layer, autonomously verifying AI subsystems by correlating multi-sensor data outputs to assess consistency. It computes a system-level Reliability Index (RI), serving as a unified measure of trustworthiness. This work introduces a hierarchical universal Meta-AI Verification Framework (UMVF) aimed at addressing limitations of singlesensor validation methods. The UMVF acts as a supervisory layer that evaluates cross-sensor consistency among AI subsystems and computes a system-level Reliability Index (RI) as a unified measure of trustworthiness. The framework is illustrated through a micro-climate monitoring case study, highlighting its feasibility and potential role in supporting trustworthy multi-sensor AI architectures. This approach strengthens validation and advances the field of Trustworthy AI.