Bad Farm Data Can Make Good AI Give Bad Advice | 35

Even a powerful AI model can give poor agricultural advice when its inputs are wrong. In this video, Professor Bannayan explains four common farm data problems: • Missing values • Sensor drift • Incorrect units • Duplicated records Using a small synthetic irrigation dataset, we show how interpreting one inch as one millimeter changes a calculated refill amount from 12.1 mm to 36.5 mm—about three times as much water. Learn why measurement units, sensor checks, metadata, and documented corrections matter before trusting an AI recommendation. This demonstration uses a simplified water balance for education. It is not a field irrigation prescription or a validated AI model. Explore AgriEnv AI: https://www.agrienvai.com Subscribe:    / @digitalagailab   #DigitalAgriculture #AgriculturalAI #DataQuality #SmartIrrigation #AgriTech