HUMAN–AI COLLABORATION IN AGRICULTURE 2035: DESIGNING A FARMER-CENTRED INTELLIGENT FARMING ECOSYSTEM

Agriculture is entering a technological transition in which artificial intelligence may become a regular component of farm-level decision-making. In 2035, farmers may interact with AI systems for crop monitoring, weather interpretation, pest identification, irrigation planning, resource management and market-related decisions. However, the future of agricultural intelligence should not be defined by automation alone. The more significant challenge is to determine how computational intelligence can work alongside human agricultural experience. This study proposes a Farmer-Centred Human–AI Collaboration Framework for 2035 (FHACF-2035). The framework considers the farmer, rather than the technology, as the central decision-making participant. AI is positioned as an analytical partner capable of processing large and diverse datasets, while farmers contribute contextual knowledge, experience, values and practical judgment. The research focuses on six dimensions: perceived usefulness, trust, explainability, farmer autonomy, accessibility and climate-resilience value. The study adopts a descriptive and exploratory research design supported by primary questionnaire data and secondary evidence. A proposed sample of 300 agricultural stakeholders is used for the empirical component, with farmers forming the major respondent group. A five-point Likert scale is proposed for measuring perceptions, while descriptive statistics, reliability analysis, correlation, regression and group-comparison techniques are proposed for data analysis. Current evidence indicates that agricultural AI is already moving into practical deployment. India is using AI for farmer advisories, pest surveillance, crop identification, insurance and agricultural governance. The National Pest Surveillance System, for example, uses AI and machine learning and is being used by more than 10,000 extension workers. Internationally, FAO and the World Bank are emphasizing responsible, inclusive and context-specific agricultural AI rather than technology deployment alone. The study argues that the agricultural AI of 2035 should follow a “recommend–explain–evaluate–decide–learn” cycle. Such a model can combine computational capability with farmer knowledge and create an agricultural ecosystem that is more transparent, accessible, resilient and human-centred.