Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Abstract. Generative Artificial Intelligence (Gen-AI) can enhance human-computer interaction by enabling natural, adaptive, and personalized communication to significantly improve data integrity and quality through automation and flexibility. Meta Ag, an Android-based farm recordkeeping app was developed to address challenges of error-prone manual data entry and it automatically recorded spatial and temporal information of farm operations; it employed a rule-based chatbot with data validation mechanisms to document operational details. In this work, the capabilities of Meta Ag were extended by incorporating a retrieval augmented agentic-AI framework to facilitate precise and complete metadata recording. The designed framework utilizes Gen-AI capabilities through API to collect and validate metadata from a user maintaining cognitively aligned human-AI interaction. The framework uses the data storage backend as a retrievable knowledge base and creates contextually appropriate questions for the user about detailed farm activity, replacing the rule-based chatbot in Meta Ag app. Implementing this framework, an example web application, Meta Ag 2.0, was developed where ChatGPT (GPT-4o) serves as the agentic AI engine. The app automatically transmits recorded spatial and temporal data to the AI, which applies procedural memory to query a private Airtable database (knowledge base, backend) to access pre-defined metadata fields. The AI dynamically generates context-specific questions, validates user inputs against database options, and refines metadata accuracy through iterative interaction before writing data records. This AI-enhanced recordkeeping system minimizes errors, reduces complexity, and enables the seamless documentation of diverse data through robust, scalable and human-centric data management solutions. This approach enables richer contextual data to support digital agriculture advancements.