The World Food Programme (WFP) estimates that approximately 343 million individuals in 74 countries are facing acute hunger, with 1.9 million individuals experiencing catastrophic hunger. In 2022, 2.4 billion people globally endured moderate to severe food insecurity, and FAO estimates that 582 million individuals worldwide will be malnourished by the end of this decade, with Africa accounting for roughly half that number. These combination of external factors such as geopolitical conflicts, trade wars, migration, global inequality and the risk of economic recession have prompted the United Nations to proclaim food insecurity a global crisis. As of April 2025, food prices had risen 7.6 per cent from the previous year, with domestic food price inflation remaining considerably high in low-income countries. Restrictions on global food and fertilizer trade, forecasts of economic stagnation and the risk of extreme weather events such as droughts, floods and severe storms (along with a projected global temperature rise of 1 to 2°C) have made the prospect of achieving SDG 2 (zero hunger) increasingly unlikely. Indeed, global food demand is expected to increase by 35 per cent to 65 per cent between 2010 and 2050, putting an additional strain on scarce resources, and threatening the food security of millions.
The incorporation of AI into geospatial analysis has helped address and mitigate some of these risks. Through the use of satellite imagery, drones, remote sensing and deep learning models, GeoAI can equip farmers and decision makers with the tools needed to build resilient and sustainable agricultural systems. GeoAI tools have provided instantaneous and granular insights into farming processes through the monitoring of crop health, soil quality, weather patterns, water usage and pest outbreaks. Furthermore, satellite imagery, drones and connected devices have made it possible to employ high-resolution imagery to track crop health in all weather conditions, augmented with field imagery to provide real-time assessments and additional layers of data for analysis.
GeoAI has also been utilized to address critical challenges such as inequity, waste and resource management by developing climate-resilient crop varieties that can withstand drought, heat or salinity. This can serve as a safety net for farmers, decreasing the stress and financial burden caused by unpredictable climate disruptions. For example, the Government of Uganda has used statistical models that analyse ground observations and satellite-based data to assess historical drought-induced crop failures and establish predetermined hazard thresholds to scale up disaster risk finance. In India, startups leverage AI and on-farm sensors and devices to deliver targeted insights to farmers based on their specific geography, crops and stages of agricultural development, lowering costs and improving crop yields. Additionally, FAO’s geospatial platform, Hand in Hand, leverages satellite-derived analytics and data to contribute to rural development and to drive digital agricultural transformation. These applications are part of FAO’s broader initiative to harness geospatial technologies for early warning systems and resilience building in agriculture.
In the field of agriculture, autonomous drones and robotic harvesters can be used to optimize crop yields and reduce resource consumption, thereby contributing to global food security.94As improvements in batteries and fuel cells continue, alongside the miniaturization of LLMs and large visual models (LVMs), coming years will see rapid developments in robotics that can reduce the barrier of entry and improve human-machine interaction.
AI for Good in focus: GeoAI and the digital transformation of agriculture, water and food systems
On 21 September 2022, the AI for Good platform conducted a webinar showcasing GeoAI’s critical role in enhancing sustainable agriculture, water management and food security. The webinar was delivered by Yanbo Huang, Research Agricultural Engineer at the United States Department of Agriculture’s Agricultural Research Service, Genetics and Sustainable Agriculture Unit in Mississippi, and Zhongxin Chen, Senior IT Officer in the Digitalization and Informatics Division at FAO. The speakers explored the evolution of agricultural information technology and its use in precision to guide smart agriculture, focusing on the increasing use of ML, deep learning with explainable AI, reinforcement learning for crop control, and generative adversarial networks for image augmentation. The presenters noted how AI is used for real-time decision-making, with unmanned aerial vehicles (UAVs), satellite imagery and IoT sensors monitoring crop health, soil conditions and water-use efficiency. The speakers emphasized the need for developing interoperable AI models which can convert data driven insights into actionable and effective farming techniques, hence increasing trust and usability.
The webinar also highlighted the research and applications of GeoAI across various regions and scales, including FAO’s Hand-in-Hand Geospatial Platform, which standardizes data for global agricultural monitoring. It demonstrated how combining multiple data sources can support comprehensive agro-environmental monitoring to boost productivity, profitability and sustainability in both agricultural and water resource management. The speakers also discussed the application of generative adversarial networks and diffusion models for augmenting agricultural imagery to address challenges such as overcoming data scarcity in training AI models for pest and disease detection, yield forecasting and responding to global issues like climate change and labour shortages.
Furthermore, the webinar illustrated how AI enables damage assessments such as post-hurricane crop loss evaluations, and supports water productivity mapping using tools such as FAO’s WaPOR platform. It encouraged open innovation through initiatives such as ITU’s GeoAI Challenge, and emphasized the significance of ethical AI deployment aligned with the principles of transparency and equity. The webinar emphasized the opportunity provided by cloud-based tools and affordable UAVs to help remove obstacles to data access and foster greater inclusivity. Through integrating domain-specific agricultural expertise with cutting-edge AI technologies, the webinar outlined a roadmap towards a more scalable, climate-resilient agriculture that empowers stakeholders ranging from researchers to policymakers.
Suggested citation: Agriculture and food security : UNU-CPR, 2026.
