Singapore Vertical Farm AI System Achieves Fully Automated Planting, Yield Up 40%
On July 30, 2026, the National University of Singapore (NUS) and Temasek Holdings jointly announced that their co-developed AI planting system had completed trials at three local vertical farms, achieving fully automated planting management, increasing average vegetable yields by 40% and reducing energy consumption by 25%. This breakthrough marks a new stage in AI application in precision agriculture and provides key technical support for Singapore, a city-state, to achieve its goal of 30% local food supply.
How Does AI Make Farms "Unmanned"?
Called "SmartFarm AI," the system integrates computer vision, IoT sensors, and deep learning algorithms. At the pilot farms, hundreds of sensors monitor temperature, humidity, light, CO2 concentration, and nutrient solution components in real time. The AI model analyzes data every 5 minutes and automatically adjusts LED spectra, irrigation frequency, and ventilation intensity. More critically, the system uses image recognition to issue warnings 24 hours before plants show disease or deficiency symptoms, reducing disease loss rates from 8% under traditional methods to below 1%.
Project lead, NUS Engineering Professor Chen Zhiming, said: "We trained the AI on over 2 million plant growth images, enabling it to judge each plant's 'health status.' It can even dynamically adjust growth rates based on market orders – for example, if an order requires larger lettuce, the AI will appropriately extend light duration and adjust the nutrient solution formula." Currently, the system can grow over 30 types of leafy greens, herbs, and microgreens, requiring no human intervention throughout the entire growth cycle.
Commercial Deployment and Cost Advantages
A senior executive at Temasek's agri-tech subsidiary revealed that the three pilot farms have achieved stable operations for six consecutive months, with per-kg production costs 28% lower than traditional LED vertical farms. By end of 2027, SmartFarm AI is expected to deploy 20 systems in Singapore, covering about 15% of local vegetable output. Additionally, the system has secured orders from UAE, Japan, and Australia for desert and urban farming scenarios.
Notably, the system adopts a modular design; a single container-sized cultivation unit can operate independently, making it ideal for land-scarce urban environments. Each unit can produce about 50 kg of vegetables per day, consuming only 120 kWh of electricity, equivalent to the energy use of an air conditioner.
Significance for Singapore's Food Security Strategy
Singapore currently imports about 90% of its food. The government has set the "30·30 Vision" – by 2030, local agricultural products will meet 30% of nutritional needs. An official from the Singapore Food Agency stated that the AI planting system is a key technology to achieve this goal. Due to year-round high temperatures and heavy rainfall, outdoor farming is inefficient, while indoor vertical farms face high energy and labor costs. AI automation precisely addresses these pain points, enabling vertical farms to achieve annual yields per square meter over 10 times that of outdoor farmland.
Additionally, the system can adjust production plans in advance based on weather forecasts and port logistics data. For example, when AI predicts that Malaysia's rainy season may reduce lettuce imports, it proactively instructs the farm to increase production of alternative varieties, thereby stabilizing market price fluctuations.
Investor Perspective: The Rise of AI Agriculture Concept Stocks
This news also caught the attention of capital markets. On the Singapore Exchange, stocks of agri-tech related companies saw abnormal movements. Among them, a local electronics company (code: S68) providing sensor modules for SmartFarm AI saw its stock rise over 15% in a single week. Analysts believe that as AI agriculture solutions move from pilot to scale, related equipment manufacturers, algorithm developers, and system integrators will enjoy growth dividends. They suggest investors focus on vertical farming companies with proprietary AI algorithms and actual orders, as well as companies providing chips and communication modules for agricultural IoT.
Notably, the machine learning model used by SmartFarm AI is not a "black box" – it can explain the logic behind each decision (e.g., "why increase light"), making it easier to pass food safety certifications. In the future, the system may also incorporate generative AI, allowing farmers to obtain planting advice through natural language conversations, further lowering the barrier to use.
Overall, the success of SmartFarm AI is not only a technological breakthrough but also an innovation in agricultural production models. It has pushed Singapore to the forefront of AI-powered urban agriculture and provides a replicable model for global food security.
