DEVELOPMENT OF A CONCEPTUAL AI-BASED DATA VERIFICATION TECHNIQUE FOR SMART CONTRACTS

Smart contracts have transformed digital transactions by enabling autonomous and transparent execution of agreements. However, smart contracts’ reliability is constrained by persistent challenges such as code vulnerabilities, logic flaws, and inaccurate input data collected from reality capture technologies, including drones, IoT sensors, and LiDAR. Conventional verification techniques primarily address code and logic, but such techniques lack effective mechanisms for input data verification, leaving smart contracts exposed to errors that may cause disputes, delays, and financial losses. The study aimed to develop and evaluate a conceptual framework for artificial intelligence (AI)-based verification technique for smart contract input data. Design Science Research Methodology (DSRM) was adopted, following its structured steps to guide the development and evaluation of the framework. The findings demonstrate that challenges remain in verifying input data, AI integration is driven by the need for accuracy, efficiency, and reliability. AI tools such as machine learning and graph neural networks were identified as suitable for input data verification. A schematic diagram was first designed to outline system interactions, which then guided the development of the conceptual framework. Expert evaluation through a focus group confirmed the framework’s clarity and feasibility while underscoring the importance of pilot testing to validate effectiveness in practice. The study concludes that AI-driven verification can significantly enhance smart contract reliability by improving data accuracy, reducing disputes, and fostering greater trust among stakeholders. The findings confirm earlier studies on the limitations of conventional verification approaches that focus mainly on code and logic, while shifting the emphasis toward AI-based techniques capable of verifying external input data. By addressing this unresolved gap in prior research and strengthening the verification of input data derived from reality capture technologies, the study provides a pathway toward more reliable, efficient, and widely adopted digital agreements.