California State University, Fresno and the Small Business Development Center hosted a Global AgTech & AI Innovation Summit examining the opportunities, challenges, and immediate applications of artificial intelligence technologies within the Central Valley agricultural industry.
The opening panel from the May 26 summit featured Ryan Dinubilo, Director of Innovation at F3 Innovate; Eric Hadden, Director at Water Energy and Technology Center; Olivier Jerhagnon, Founder and CEO of AgMonitor; and Tiffany Weldin, Manager of Data Analysis at Primex Farms.
The technology vendors largely described AI in broad strategic terms, emphasizing that recent advances in computing power and declining hardware costs are finally allowing artificial intelligence systems to move from theory into practical deployment.
A recurring theme throughout the discussion was that the primary bottleneck to adoption may no longer be technology itself, but workforce readiness and trust.
Panelists noted that many growers remain cautious about relying on what are often perceived as “black box” systems — technologies that generate recommendations or decisions without easily understandable explanations of how conclusions are reached.
Speakers also discussed the fragmented nature of current agricultural AI deployments. Many systems operate within isolated silos — such as pest detection, irrigation monitoring, or equipment management — without full integration across operations. Some participants suggested that local data centers and regional computing infrastructure may eventually become necessary to support large-scale agricultural large language model (LLM) processing while keeping sensitive operational data closer to growers.
The presenters also emphasized the importance of “reference-ability” within agriculture, where successful technologies often spread grower-to-grower through direct observation and neighbor recommendations rather than through marketing alone.
Among the most concrete examples presented came from Primex Farms’ pistachio processing operation in Wasco.
According to Weldin, the company is implementing AI technologies in phases and is currently operating its next generation of deployment systems. The representative reported that AI-assisted optical sorting technology has tripled processing speed, allowing the facility to process approximately one full truckload of pistachios per hour.
The system continuously learns product characteristics and automatically adjusts equipment settings to maintain consistent quality despite variations in incoming raw materials. Primex Farms plans to further expand its use of AI-driven statistical analysis and production forecasting capabilities.
The panel was followed by a presentation from Dr. St. Hilaire, Dean of the Jordan College of Agricultural Sciences and Technology.
One highlighted technology involved precision “spot spray” systems using computer vision to identify weeds and apply herbicides only where needed. According to the presentation, the system has demonstrated herbicide reductions of up to 90 percent compared to conventional spraying methods.
Initially introduced as rental equipment, the technology reportedly proved successful enough that manufacturers are now scaling production for direct purchase by growers.
Dr. St. Hilaire also described “bolus” monitoring technology, from smaXtec, being deployed within Fresno State’s dairy operation. Approximately 90 animals have been equipped with ingestible sensors — each only slightly larger than a small flashlight — capable of monitoring rumination, movement, body temperature, and water intake.
The system can issue alerts tied to potential health concerns and detect estrus cycles through body temperature changes as small as one-half degree Fahrenheit, allowing breeding windows to be identified within roughly 24 hours. The vendor claims it can help dairy farmers reduce the use of antibiotics by up to 70 percent.
Additional presentations explored hyperspectral imaging technologies capable of analyzing plant conditions beyond the visible light spectrum. Further research described systems that can identify water stress, nutrient deficiencies, and plant disease before symptoms become visible to the human eye.
One example presented involved Pierce’s disease detection in grapevines. Researchers stated that hyperspectral analysis identified infection across approximately 87 percent of a vine while only about 3 percent of visible symptoms had emerged, potentially creating a wider treatment window for bacteriophage therapies.
Dr. St. Hilaire also announced that Fresno State’s AI supercomputing center is now fully operational and available to provide local computing resources for agricultural applications. The initiative is intended, in part, to address grower concerns about sending sensitive operational data to distant cloud-based processing centers.
If the summit revealed anything clearly, it is that AI adoption within Central Valley agriculture is no longer hypothetical. The remaining questions may be less about whether these systems will arrive, and more about how quickly growers learn to trust them — and who ultimately controls the data, infrastructure, and decision-making systems behind them.