Artificial Intelligence

Albertsons Debuts AI Tool to Boost Produce Quality

Thu, 14 May 2026 20:05:47 GMT

Albertsons Cos. Inc. says it has launched its patent-pending Intelligent Quality Control tool, an artificial intelligence-powered solution that uses computer vision to support the ability of distribution center associates to maintain high standards for quality and consistency of fresh fruits and vegetables across the company’s supply chain.

Built in-house by Albertsons Cos. technology and supply chain teams, the tool uses Google Cloud’s Gemini Enterprise, including Vision AI and Gemini models, to support Albertsons Cos.’ quality inspection teams to help ensure customers consistently receive fresh, high-quality produce.

“We built [the] AI-powered Intelligent Quality Control tool to support our team of talented quality inspectors in our distribution centers, and early results show it’s been incredibly helpful in increasing the consistency of quality rating, which is crucial for highly perishable products such as fruits and vegetables,” says Evan Rainwater, executive vice president and chief supply chain officer for Albertsons Cos. “This is just the latest advancement in how we are using AI within our multibillion-dollar supply chain to improve operational efficiencies, improve product quality and ultimately enhance customer satisfaction.”

The Intelligent Quality Control solution was built in-house by Albertsons Cos. teams with advisory support and technology infrastructure from Google Cloud as an enterprise partner. Seamlessly integrating into the distribution center workflow, the quality inspector feeds an image of the produce into AI tool, powered by Google Cloud’s Gemini Enterprise Agent Platform, which evaluates visual characteristics against Albertsons Cos.’ established quality standards.

The system then provides a highly accurate, consistent rating and recommendation to the inspector for approval. By reducing ambiguity and enabling more data-driven decision-making, the Intelligent Quality Control solution provides valuable support to help distribution center teams achieve greater accuracy, consistency and confidence in their work, according to the company.

Live now in select Albertsons Cos. distribution centers, the tool is initially focused on quality inspections for strawberries and red and green grapes. The company is expanding the solution across the entire berry section with plans to scale the tool nationwide and incorporate more fresh products. Albertsons says early results have demonstrated significant operational improvements including:

Increase in rating consistency — AI-driven evaluation has reduced the variability in quality ratings among different inspectors and shifts at a location, ensuring a more uniform standard.
Faster decision-making — The automated scoring system accelerates the inspection process, supporting distribution center teams in getting food quickly to stores so that customers consistently receive fresh, high-quality produce.
Expanded quality data — The solution captures many distinct measures of quality that can be used for ongoing quality analysis and improvement.
Alignment with standards — The solution successfully applies internal Albertsons Cos. quality criteria, with high consistency across all evaluations.

“This collaboration with Albertsons Cos. demonstrates the transformative power of applying advanced AI, like Vision AI and Gemini, to the core of the supply chain,” says Jose Gomes, vice president of retail and consumer packaged goods for Google Cloud. “Ensuring quality consistency for fresh produce is a complex, logistical challenge. By advising on this intelligent component, we are helping Albertsons Cos. drive efficiency and, most importantly, deliver on their promise of fresh, high-quality food to their customers. This is the future of agentic commerce.”

How Artificial Intelligence Is Transforming the Fresh Produce Supply Chain

Mon, 04 May 2026 11:32:34 GMT

TORONTO — A panel at the Canadian Produce Marketing Association Convention and Trade Show discussed both the opportunities and the challenges of using artificial intelligence in the fresh produce industry. While moderator Steve Roosdahl, CEO and president of BC Fresh, warned that AI can hallucinate and create false information if not fed good data, the benefit to closed data sets was discussed by panelists Stewart Lapage, vice president of supply chain and logistics for The Oppenheimer Group; Mike Meinhardt, North American business development executive for Clarifresh; Tim Raiswell, CEO of Oxrow.ai; and Alex Carvalho, chief technology officer of Bloom IQ.

“AI is kind of like your smartest friend that does hallucinogenic drugs. If AI doesn’t know, it makes it up,” Roosdahl says. “But as it learns, it needs feedback, and it needs good data to make good decisions.”

Meinhardt says these closed data sets, which use only the collected information the company supplies, ensure analytics are accurate.

“That’s a very healthy place to be,” Meinhardt says. “You don’t have to worry about garbage data because it’s your data — your data only that you’ve collected.”

Carvalho says that grounding models using specific searches and algorithms ensure information users act upon is consistent and accurate.

Scaling Operational Accuracy From Packinghouse Sorters to Global Logistics

When asked about the potential benefits AI can present, Raiswell says it falls in two categories. The first is acceleration of work, which includes automating repetitive, standardized tasks like scheduling, processing invoices and compliance. The second includes the ability for organizations to perform tasks previously impossible without specific skills.

Meinhardt says that AI can help scan fruit much faster within packing operations and screen for specific defects.

“We use AI to measure size, color, color coverage, stem color and external defects,” he says. “[We use] machine learning to identify cracks versus scars versus sunburns and so forth.”

Roosdahl says AI can offer consistency that might not be possible with even the best sorter.

“You get consistency because you get a machine that — when people get tired, they miss things, but machines tend not to,” he says.

Lapage says he sees a huge potential for AI’s use in farming to boost long-term sustainability.

“I see the real benefit being actually on the farming and growing side,” he says. “There are so many farmers and growers around the world on a knife’s edge, as we like to call it, whether their businesses and their farming operations are going to be sustainable from year to year. They can use this technology to find efficiencies, save costs, improve forecasting, yield — all the different things to make the farming side of the business sustainable.”

Roosdahl points out that drones could scout for pests and water issues and machines could automatically weed.

Lapage says, though, that data security is a huge focus for Oppy as it integrates new technology. Roosdahl cautioned the audience to ensure that data protection is a huge component of the integration as new technology is brought into an operation.

While it might seem tempting to use a public AI model that’s available — such as ChatGPT, Microsoft Copilot, etc. — Roosdahl cautions against that, as that data will be shared as common knowledge.

“If you just go with public AI and you take all your policies in your company and throw them out there, now everybody has your policies,” he says.

Evolution Toward Specialized Language Models and Proprietary Data Security

Raiswell says that the future of AI is less about the technology and more about how companies will integrate it.

“If you’re wondering how AI is going to impact my workforce, that’s more to do with you and how you structure roles than it is to do with AI,” he says.

Raiswell says industry leaders will be willing to test AI against the best human experts.

“The companies that are willing to ask that question of our gala guy, our Honeycrisp guy who knows the market better than anyone — are they willing to put them up against AI and really test that hypothesis?” he says.

Lapage also says he expects the conversation around AI at CPMA’s 2027 event in Vancouver, British Columbia, will be markedly different, but he warns that produce industry businesses that are reluctant to enact AI might not recover as technology continues to evolve at a rapid pace.

“If we don’t get on board, we’re probably going to get left behind to some degree,” he says.

Meinhardt says he sees opportunities with warehouse and operations, noting that there’s a strong potential for food safety.

“Food safety documentation [is] all rolled up into AI, giving you a food safety scorecard and telling you what your risks and analysis [are] and then, you know, trying to send the alarm light before the alarm light needs to be sounded,” he says.

Carvalho says that AI will shift within the next five years from general large language models designed for broad tasks — such as ChatGPT, Claude and Microsoft Copilot — to smaller language models designed for very specific tasks.

“Each model knows one specific part of the process to help you optimize your systems, your day-to-day and how you work,” he says.

And he expects this small language model will also help improve the work-life balance of its users.

“In five years, I believe we will have a better work-life balance and also a more improved work with all this decision making,” he says.

Share-ify Appoints AI Product Manager and Solutions Architect

Wed, 22 Apr 2026 22:01:35 GMT

Share-ify has named Joh Johannsen its artificial intelligence product manager and solutions architect, a strategic addition that the company says reflects its continued investment in building a smarter, more intelligent platform for the food industry.

In this role, Johannsen will help lead the development and execution of Share-ify’s AI product strategy, with a focus on turning complex supply chain data into more actionable insights and more automated workflows for customers, according to a news release. The Orlando, Fla.-based company says the appointment supports its broader commitment to incorporating AI across the Share-ify platform in ways that improve visibility, efficiency, traceability and decision-making for food manufacturers, distributors, retailers and other supply chain partners.

Johannsen is a graduate of Harvard University, where he studied applied mathematics and computer science. Share-ify says he brings a strong technical foundation and engineering experience spanning both high-growth startups and major global organizations, including Journal Technologies, Amgen and Amazon.

His background combines academic rigor with practical product and engineering expertise, positioning him to help translate advanced AI capabilities into scalable tools that solve real operational challenges, the company says.

“Joh’s arrival is an important step for Share-ify as we continue expanding the intelligence and automation built into our platform,” says Ernesto Nardone, CEO of Share-ify. “We are committed to applying AI in ways that create meaningful value for our customers, whether that means improving decision-making, reducing manual work or helping companies respond faster to the demands of an increasingly complex food supply chain. As a fellow math and computer science graduate with a specialty in artificial intelligence from McGill University, I am especially pleased to welcome another mathematician to the team.”

Johannsen has been working with Claude Code since its initial release and has extensive experience across a broad range of AI and large language model tools. Over the course of his career, he has helped experiment with, deploy and refine real-world AI applications across multiple environments and use cases.

Share-ify says this hands-on experience will support its efforts to accelerate AI innovation responsibly and practically, ensuring new capabilities are aligned with customer needs and built for real-world adoption.

This hire signals Share-ify’s ongoing focus on the future of food safety, compliance and supply chain collaboration through technology that is more predictive, more responsive and easier for customers to use, the company says.

As Share-ify continues to evolve its platform, it says AI will play an increasingly important role in helping customers manage product data, identify risks earlier and operate with greater confidence and speed.

Farmers Don’t Use AI for Answers — They Use It to Think Better

Fri, 17 Apr 2026 15:10:21 GMT

What you should know:

To use artificial intelligence in your business for a competitive advantage — not just a gimmick:

Ask better questions than most people
Combine AI with real-world experience
Execute on the answers

For Rachael Sharp, dry weather hasn’t made planting go any easier in the Lowcountry of South Carolina. And when a planter went down, the first thing she did was pull up Chat GPT.

“I pulled up the part number, and I saw that I’d actually entered in there last year. So it told me the date I changed it, and that was helpful, because I was trying to figure out why is this wearing out so quickly?” she says. “We’re in desperate need of rain, and we’re pulling in some pretty hard non-irrigated land right now. I logged that we changed the bearing again, and so next time, knock on wood, it hopefully doesn’t go out again, but if it does I can look and see I changed it twice in the last year.”

That’s just one of many examples of how Sharp is using ChatGPT to manage equipment, her time, and the farm business. She and her father, Don, are featured in an OpenAI commercial, which premiered during the Super Bowl.

And she’s in good company with other farmers in how to use the artificial intelligence platforms.

Marc Arnusch, the 2025 Top Producer of the Year, says ChatGPT is the most used app on his phone.

Jeremy Jack, leader of Silent Shade Planting Company the 2023 Top Producer of the Year, uses AI as his daily management teammate from agronomy and business decisions.

Here are the four ways these farmers use AI every day on the farm.

(Photo: Yvonne Min)

1. Make better decisions faster

Colorado farmer Arnusch uses ChatGPT and Grok to narrow down his consideration set when making decisions.

“It helps on the strategic side of things, and when making a decision, I’ll let it give the top four or five things to choose from, which helps when there’s a million choices,” he says. “It really is like my funnel. I’ll set up my phone on my dashboard and just dictate to it. Then when I’m back at the farm office, my wife Jill is relieved because I’ve already processed out loud with the AI tool.”

While most farms collect data, Jack uses AI to make decisions, particularly agronomic.

“I uploaded multiple years of soil data across our farms,” he says. “And we’ve found ways to manage fertilizer better, for example with sulfur.”

The data interpretation has shifted his thinking by connecting the yield zones with as-applied fertility and return on investment.

Jack is also using the technology to double check every spray application — from rates, to tank mix, to nozzle selection, to pressure optimization.

Sharp has also found AI helpful in managing chemical applications. She can remember chemical boxes marked up with her father’s calculations by hand.

“I tell the prompt what I’m spraying, where I’m spraying, how many acres, tank size, and then I let it tell me what to order,” she says. “Over time, it’s learned which products are liquid and which are dry flowables. And it’s helped me keep track of the inventory we have so we don’t end up with pallets of odds and ends.”

(Photo: OpenAI)

2. Be more efficient

When it comes to where to start with AI, Sharp has one piece of advice.

“Think of the task that you don’t like to do at the end of the day. For me, I didn’t want to do paperwork at the end of the day,” she says. “So I threw it over to ChatGPT, and I said, hey, this is what I planted today, this is the date, and I left it at that. I started really, really simple.”

Now, she’ll record things directly in the field or in the truck. She says it has helped with FSA 578 forms. And in day-to-day operations, she’s found benefits for time management and accuracy in all record keeping.

“We have seed samples that require a handwritten seed form that I turn in along with the sample, but I spoke into my phone and said, hey, Chat GPT, I need you to log that I sent this variety, this lot number, on this date, to the lab. And so, that’s probably one of 15 entries that I’ve made over the course of a month. And at the end when we finally turn in our last sample to the lab, I’ll ask it for a spreadsheet with all that listed,” she says.

(Photo: Sarah Green Photography)

3. Think more clearly about complex problems

Jeremy Jack often asks ChatGPT “What does this mean for my farm?” with current events.

“With the war in Iran, global fertilizer supply chain concerns, and even things like USDA reports, it’s given helpful perspective in how to think about what’s happening off the farm but impacts the farm.”

And he’s found success in using the platform to specifically think about the business strategy for his farm with vendors, including lenders, landowners and more.

4. Manage more professionally

Jack has been active with an advisory board for their farm, but AI has become like a boardroom in his pocket.

“I bounce ideas—pressure test if you will—before it costs me real money,” he says. “This includes input purchases, land agreements, and equipment purchases.”

He’s also come to use it in his external communications about the farm including his regular social media posts on LinkedIn.

When it comes to team management, Arnusch has input culture index results for vendors and employees, then the AI compares their individual characteristics with the job they are being asked to do.

“This has been a breakthrough,” he says. “It’s shown me that at no fault of their own, why some people fail at what they are being asked to do. It wasn’t because they weren’t working hard or doing the job. It was stretching them beyond what they can do.”

He gives the example of a farm foreman position on the farm, and how he used this process to match the candidate with the role.

(Sarah Green Photography)

The New Ag Economy: Why This Downturn is a Structural Shift, Not Just a Cycle

Mon, 13 Apr 2026 21:22:20 GMT

What You Need to Know:

Structural Evolution: This downturn is a permanent market shift, not just a temporary cycle.
Friend-Shoring: Trade is moving toward geopolitical allies to ensure supply chain resilience.
Aggressive Cost-Cutting: Farmers are doubling generic input use and delaying machinery purchases to protect margins.
Financial Resilience: Better management and working capital make today far more stable than the 1980s.
Premium Protein Demand: GLP-1 medications are driving consumers toward smaller, higher-quality meat portions

As the industry enters the third year of this downturn, farmers and agribusinesses are questioning if a recovery is on the two-year horizon. While cyclical behavior is normal, two economists suggest the structural evolution within crop protection, machinery, technology, livestock and other individual sectors is creating a different kind of staying power for those who survive the recovery.

The Evolution of the Cycle

When characterizing the current economic cycle in agriculture, historical patterns provide a necessary baseline, yet the present landscape is defined by unique pressures. Typical agricultural cycles consist of roughly six years of expansion followed by four years of decline. Currently, the market is navigating a “corrective period,” returning to long-run averages.

The drivers of growth are typically demand shocks — export surges, fuel demand or policy shifts such as the Renewable Fuel Standard. However, Wes Davis, ag economist at Meridian Ag Advisors, notes the current environment is an intersection of traditional contraction and sector-specific evolution.

“What I think we’re experiencing right now is that typical cycle behavior where we see growth in some business firms, and then some contraction and pullback to adjust to the cycle going back to more of the long-run average,” Davis explains. “I think we’re also seeing evolution of individual sectors within the market where there’s adjustments happening because of the industry itself.”

In other words, this isn’t just a cycle — it’s also a structural shift.

Change Fatigue and Modern Volatility

Farmers aren’t strangers to volatility, but global trade disruptions, policy shifts and rising competition, especially from Brazil, are layering uncertainty onto already volatile markets.
Farmers are grappling with “change fatigue,” a byproduct of the high velocity of information and extreme price swings that dwarf the relative stability of the early 2000s.

“When I go talk to any industry group right now, the phrase that I hear is ‘change fatigue’, and I feel that. Every couple minutes, something shifts,” says Trey Malone, Purdue University ag econ professor. “But to be clear, it’s not that the farm economy isn’t used to volatility, it’s just the uncertainty and the volatility now is, like, ‘hold my beer relative’ to the old volatility.”

Malone attributes this to layers of uncertainty created by global trade and policy. The rise of Brazilian production, coinciding with the disruption of U.S.-China trade relations, has created a permanent state of flux. This sentiment is reflected in the Purdue Ag Economy Barometer, which shares a higher correlation with the Small Business Index (.5) than with actual commodity prices. This suggests farmers view themselves primarily as small business owners facing broad economic pressures rather than just price-takers.

“We don’t see very strong correlations even with lagged soybean prices and corn prices,” Malone notes. “The world is more complicated than just looking at what happened in the market yesterday and gauging how farmers feel.”

Global Competitiveness and the Trade Reallocation

A primary concern for U.S. producers is their position as low-cost providers. While the U.S. maintains an infrastructure advantage that lowers the cost of getting products to export ports, Brazil continues to close the gap.

“It’s a fair question farmers ask a lot: Are we actually the ones who are the low-cost producers, and do we still have a place in the global market if Brazil continues to lower the cost of production and transport their grain to export terminals?” Davis asks.

However, Davis points out that global trade hasn’t shut off; it has reallocated. Only three global regions — North America, Latin America and parts of Southeastern Europe/Central Asia — are net exporters. The rest of the world remains net importers.

“While our trade has kind of shifted around ... that shift has really reallocated stuff in different places. Those calories and products end up going somewhere. It’s just a question of where,” he says.

The Shift to “Friend-Shoring” and Resilient Supply Chains

The industry is moving from “just-in-time” (hyper-lean) procurement to “just-in-case” (inventory-heavy) strategies, a lesson reinforced by the pandemic. This shift is accompanied by “friend-shoring,” where the U.S. prioritizes trade with geopolitical allies.

“We’ve gone from offshoring to onshoring to nearshoring to friendshoring,” Malone explains. “We’ve got a paper that’ll be coming out ... where we document friend-shoring in ag and food supply chains. Over the last 10 years, there’s been a shift where we mostly in the U.S. trade with other people who vote like us in the WTO. That’s kind of one way to measure friends.”

This resilience is also visible in crop protection. In 2019, 80% of active ingredients were sourced from China. Today, that is closer to 60%, with manufacturing shifting to India and domestic sites. Davis calls these “geopolitically resilient” supply chains.

The Rise of Generics and Decision Paralysis

The economic downturn is fundamentally changing the business model for input providers. Farmers are aggressively cutting costs, leading to a massive surge in generic usage.

“The latest survey I saw shows about 60% of farmers use generics today. That was about 30% to 40% just 5 years ago,” Davis says. This forces companies to pivot from differentiation to operational volume.

In the machinery sector, high costs and economic uncertainty have led to “decision paralysis.” Farmers are extending the life of their equipment, treating machinery replacement as the most controllable variable in managing annual ROI. Davis notes the U.S. ag equipment cycle is currently 15 to 20 percentage points lower than typical low points, driven by this hesitation. Furthermore, there is significant skepticism toward subscription-based technology models.

“Farmers don’t terribly love this idea, and I think the other interesting thought here is I’m not sure that retailers like selling them either,” Malone adds.

AI: The “Undergraduate Intern”

While artificial intelligence (AI) is a major talking point, its current role in agriculture is more supportive than transformative. Malone views AI as a “highly capable undergraduate intern” — useful for processing information but incapable of replacing the trust and risk management provided by human advisors.

“I don’t think you need to be replacing your agronomist. I think your mediocre agronomist just got OK,” Malone says, noting while LLMs can pass CCA exams, they cannot manage the risk of a wrong decision. “The risk management value proposition of an in-person Claude, or whoever, is probably going to win out because there’s still a risk.”

Currently, the adoption gap is wide: While 75% of agribusiness managers see potential in AI, only 4% have implemented it, according to a Purdue University survey in 2025.

Livestock and the GLP-1 Impact

The livestock sector is facing a unique demand shift driven by weight-loss medications (GLP-1s). This is leading to “premiumization.” As consumers eat smaller portions, they are opting for higher-quality cuts.

“The explosion in demand for protein is just shocking,” Malone says. “What GLP-1s do to that calorie count is they are all shifting toward premium cuts. You don’t care how much it costs because you’re only going to have seven bites of it. But you’re going to have a steak. That premiumization is going to really, really take off in the next 10 years.”

Conversely, the hype surrounding “fake meat” has largely faded, proving to be more of an investor-led phenomenon than a market-driven one.

Financial Stability: Not the 1980s

Despite the downturn, the financial health of the American farmer remains more stable than during the crisis of the 1980s. Currently, 10% to 12% of farmers are in a “tight” financial position , compared to 20% to 30% in the 80s.

“We do have a completely different, more professional ag workforce than we did back then,” Malone says. “The farm policy we have right now does not necessarily match what we need for the future, but all of these things make me think we’re in a much more stable position.”

Farmers have built-in “shock absorbers,” Davis adds, including off-farm income and working capital built up during the expansion years. However, in his research Davis has seen how alternative financing is becoming a major tool for the 50% of farmers who use it — either to manage stress or, for larger operations, to leverage relationships with retailers.

Strategic Reassessment: Winning at the Bottom

The experts agree the “bottom of the cycle” is the time for professionalization and upskilling. Surviving — and thriving — will require sharper management. It is an opportunity to reassess farm transitions and management disciplines, such as financial management, accounting and planning, which become critical in tight margins.

“Farmers are going to have to get smarter and get more creative with how they manage,” Malone says. “This is a good opportunity to take a step back and think about what the strategy needs to be moving forward.”

Davis emphasizes relationships are solidified during these periods: “Farmers are going to remember the folks who were around when they were in the bottom of the cycle, and who were there to support them. The best farmers will continue to get better ... I get excited about what we can look like as we come out of this cycle.”

So Is This Ag Cycle Different?

These experts say yes as every cycle presents its own unique reshaping of future opportunities.

To download the full report on why this ag cycle is different and what it means for your operation, click here .

‘Farmville for Real:’ How Autonomous Tech is Rebranding Tractor Drivers as Digital Operators

Fri, 10 Apr 2026 21:18:38 GMT

For George Grote, a typical day doesn’t involve a steering wheel or a dusty cab; instead, it looks like he’s glued to his phone or tablet. From the climate-controlled comfort of a pickup or while strolling between the crop rows, Grote monitors a fleet of autonomous tractors as they navigate with precision. It’s a scene that Agtonomy CEO Tim Bucher calls “Farmville for real,” where gripping the steering wheel is being replaced by the quick-twitch reflexes of the gaming generation.

As the agricultural industry grapples with a deepening labor crisis and an aging workforce, companies such as Agtonomy are betting that high-tech autonomy is the key to recruiting Gen Z. By rebranding traditional tractor driving as “digital operation,” the California-based startup is leveraging app-based interfaces to transform farming into a tech career.

Raised on a dairy farm, Bucher coupled his love of agriculture and machinery with a career in the tech field to create the prototype that would become the company’s first autonomous tractor. Today, as a farm owner and father of three, Bucher says it’s not likely his children will return to the farm. With this technology, he hopes to attract the next generation to his farm and the industry.

“The younger generation doesn’t want to be in the dirt and the dust because there are other opportunities for them,” he says.

(Photo provided by George Grote)

The Gaming Advantage: Why “Fast Hands” Matter in the Field

Grote, a 30-year-old field engineer with Agtonomy, does not have a farming background, but he always knew he wanted to work outdoors. He graduated from California Polytechnic State University with a degree in crop and fruit science and now works alongside Agtonomy customers. He says being able to process information on a screen while being fast with his hands is something he picked up from gaming.

“I can tell you [gaming] 100% helped me, being able to toggle between different features in the app and being quick with technology,” Grote says. “If you can play a video game, then you’re more than capable of running five, six, seven or eight tractors at once while sitting in your truck and watching them run autonomously.”

8.14.25_agtonomy-102

(Agtonomy)

Marc Di Pietra, regional service maintenance manager for Treasury Wine Estates , says the cutting-edge technology can help bridge the labor gap, provide upskilling opportunities and reduce physical demands.

“New technologies are making it incredibly dynamic and attractive for younger generations,” Di Pietra says. “With a younger workforce, I expect that evolution to accelerate. There’s a natural comfort with technology, and a willingness to challenge legacy processes, which is critical.”

Adding this technology has been game changing in the wine business, De Pietra says, and it can shape the rest of the agricultural industry as well.

“These technologies allow us to be more precise with our farming, winemaking techniques and sustainability efforts, effectively creating safer environments for our employees, reducing our emissions and producing better quality wines,” he explains.

Agtonomy_George Grote

(Agtonomy)

How Agtonomy Works

At this point, Agtonomy is being used on fruit and nut tree, grapevine, avocado and citrus operations. The company works with Bobcat , specifically the Doosan Bobcat CT 4045, for more versatile utility and maintenance tasks, and Kubota to automate narrow-track diesel tractors, such as the Kubota M5N series.

Agtonomy’s software is embedded into a tractor’s OEM control system at the factory. Outfitted with the technology, the machine can autonomously handle repetitive tasks, such as mowing, spraying and seeding. Using cameras and sensors, tractors can “see” their surroundings, allowing them to navigate rows and avoid obstacles. Sensors and data links also ensure that sprayers or mowers are working at the correct intensity and height. When an issue occurs, a notification is sent to a smartphone and/or tablet, complete with details to help the tractor decide what to do when it deviates from the original instructions.

Inside The Tax Return of Your Farm's Future

Thu, 09 Apr 2026 14:31:52 GMT

The traditional process of preparing agricultural tax returns has long been defined by manual data entry and the complex reconciliation of income. However, the integration of artificial intelligence into financial systems is ushering in a more sophisticated era of tax management. For the modern farm, the future of filing lies in a seamless pipeline where software handles the heavy lifting of data organization, leaving the high-level strategy to human experts.

Comprehensive Data Integration

The foundation of a modern tax return is the accounting system. Platforms like QuickBooks, Xero or specialized farm management software are becoming increasingly autonomous. In the near future, these AI agents will do more than simply record expenses; they will analyze them in real-time.

With direct links to bank feeds and digital invoices, AI can categorize expenditures with precision. It can distinguish between capital investments, such as machinery or land improvements, and standard operating costs like seed and fuel. This continuous synchronization means by the end of the fiscal year, the financial records are already in a format that mirrors the requirements of a tax return.

Automated Document Reconciliation

A significant portion of tax preparation involves matching — ensuring the farm’s internal records align with the documents issued by third parties. A preparer of a farm tax return may spend more time making sure all of the income is in the right box then planning to optimize the income tax level.

AI is uniquely suited to handle this high-volume verification. The system can automatically ingest Form 1099-PATR (cooperative distributions), 1099-G (government subsidies) and other Form 1099s and W-2s and verify them against recorded deposits.

If a document is missing or a figure does not match the ledger, AI identifies the specific discrepancy immediately, allowing for a targeted correction rather than a manual search through months of records.

The Role of Human Oversight

While AI provides the technical framework for the return, the final stage remains firmly in human hands. Once the software has mapped the data to the appropriate tax schedules, it produces a comprehensive draft for professional review.

This allows the farmer or a tax consultant to transition from a data entry role to a strategic advisory role. Instead of spending hours verifying line items, the human reviewer can focus on critical tax planning decisions including accelerated depreciation choices or income averaging that require professional judgment and an understanding of the farm’s long-term goals.

The result is a more accurate, defensible and efficient tax filing process. By automating the clerical aspects of the return, AI allows agricultural producers to maintain focus on their operations while ensuring full compliance with the evolving tax laws.

Agronomist in Your Pocket: How AI Is Transforming Global Pest Management

Tue, 07 Apr 2026 18:59:37 GMT

For over a decade, a dedicated team at Iowa State University has been working at the intersection of artificial intelligence and agriculture with a mission to provide farmers with the tools they need to stay ahead of an ever-changing landscape of threats. Led by Arti Singh and Soumik Sarkar, this research has culminated in the development of the PestIDBot, a sophisticated AI companion designed to act as an “expert crop advisor or extension agent in your pocket.”

By combining massive image databases with conversational AI, the team is moving agricultural protection from a reactive struggle to a proactive, precision-based science.

A Decade of Data-Driven Identification

The core of the technology lies in two specialized applications: Insect ID and Weed ID, the result of training massive AI models on staggering amounts of data. The Insect ID app has been trained on 16 million images and can identify roughly 4,000 different species, ranging from common pollinators and predators to invasive threats. Similarly, the Weed ID app utilizes 15 million images to identify 1,600 weed species, including noxious and invasive varieties.

While these models are global in scope, they have been fine-tuned specifically for regions.

“If a farmer in Iowa does a web search on a pest, they might get information relevant to the Southern U.S. that isn’t applicable to an Iowa farmer,” Singh says. By narrowing the model’s focus to local threats and incorporating management practices vetted by University Extension scientists, the tool provides personalized, actionable information tailored to the user’s specific location.

Image of spotted lantern fly egg masses on tree bark is identified in PestIDBot.

(Image courtesy of Iowa State University)

From Identification to Conversation

The true breakthrough of the PestIDBot is the integration of identification with a conversational chatbot. In the field, a farmer can take a real-time photo of an unknown insect or upload an image taken previously. Once the AI identifies the pest — even in early stages, such as egg masses — the chatbot allows the user to ask contextual follow-up questions.

“Rather than searching for a human expert while the clock is ticking, you can ask your first questions directly to the app,” Sarkar says.

Users can inquire about treatment timing, the necessity of spraying or specific management steps based on their observations. For example, if the app identifies the eggs of an invasive species like the spotted lanternfly, it doesn’t just provide taxonomic details; it can advise the user to contact specific state agencies, such as the Iowa Department of Agriculture and Land Stewardship.

Solving the Green-on-Green Challenge

Building an AI that works in a controlled lab is one thing, but the field presents chaotic variables. Sarkar notes that early models lacked the robustness to handle cases like green-on-green (pests on leaves) or brown-on-brown (pests on bark or soil) scenarios. To ensure the system is trustworthy and reliable, the team implemented strict guardrails to prevent hallucinations — where an AI confidently provides an incorrect answer.

These safety measures include out-of-distribution detection, which allows the AI to recognize when it is looking at something it wasn’t trained for (like a human face) and simply say, “I don’t know.” Furthermore, when the model is unsure, it is programmed to provide several likely options rather than a single potentially wrong identification, allowing the farmer to consult with experts using a narrowed set of possibilities.

Shown from left, Iowa State University’s Arti Singh and Soumik Sarkar are part of a team working at the intersection of artificial intelligence and agriculture to provide farmers with the tools they need to stay ahead of an ever-changing landscape of threats.

(Photo courtesy of Iowa State University)

The Global Horizon: The BRIDGE Project

The next frontier for the team is the AI Engage (BRIDGE) project, funded by the National Science Foundation. While insects and weeds are well documented, identifying crop diseases caused by bacteria, viruses and fungi is a much tougher problem due to the limited quality and expert-verified data.

By partnering with researchers in Australia, Japan and India, the team is building a global dataset of disease images. This international collaboration is critical for biosecurity.

“Threats emerging in Africa or Asia will eventually show up on our shores,” Sarkar warns.

By training models on global data now, U.S. farmers can be prepared for future threats before they arrive, shifting the agricultural industry from a reactive stance to a proactive one.

Vision for Sustainable Stewardship

Beyond technical identification, the team is driven by a passion for sustainability and the future of the agricultural workforce. By enabling precision-based farming, the PestIDBot can help farmers pinpoint exactly which part of a field needs treatment. This hyperprecise approach reduces the need for blanket chemical spraying, lowering input costs for farmers while protecting water systems and overall environmental health.

Finally, Singh and Sarkar are using this technology to make “agriculture cool again” for the next generation. Through workshops and gamified modules for K-12 and 4-H youth, they are fostering land stewardship and encouraging young people to see themselves as future innovators in both ag and AI. As Singh reflects, empowering a kid in a front yard to identify an invasive species can be the first step in a statewide defense against agricultural threats.

Innov8.ag Turns Harvest Data Into a Morning Playbook

Wed, 18 Mar 2026 23:28:10 GMT

Innov8.ag has released its HarvestReplay, a service that transforms data into a decision-making tool. It looks at an operation’s labor, crop production and harvest management and provides daily, weekly, monthly and seasonal insights.

The company says HarvestReplay delivers these insights into online visualizations and customized daily audio briefings with three key features:

Replay history — Deep, retrospective analysis in geospatial views that turns multiyear harvest and labor records into reports and benchmarks showing true cost per unit, block and variety performance and the economics of past harvest decisions.
Replay live — Same‑day and in‑day feedback layered on GPS labor tracking, with benchmarks and alerts to flag issues like station congestion, slowdowns or misallocated crews so managers can reassign in the same shift.
Replay podcast — An artificial intelligence-generated, multilingual, interactive private audio intelligence briefing built from a grower’s own harvest data, tailored for each key farm role, delivered daily, weekly, monthly or seasonally to match operational needs.

HarvestReplay is built on more than a decade of commercial harvest data, which includes 12.3 million recorded farm labor hours and 11.1 million captured GPS events, according to Innov8.ag. The service also integrates proprietary and public data sources, such as university research, labor modeling and aggregated benchmarks from technology, research and industry partners.

HarvestReplay can also connect field data with packinghouse outcomes to create a closed-loop system to better inform how a crop was harvested, how it graded and performed postharvest.

‘Remember When?’ With Context

Innov8.ag founder and CEO Steve Mantle says HarvestReplay takes the historical data of an operation and provides data and context in a more modern “Remember when?” that growers so fondly speak of.

Mantle says this daily audio briefing can provide information such as the forecast ahead so growers can better anticipate potential weather interruptions, trends and context around production. He says this is all with the intention of helping growers better understand current trends and also places to improve.

Instead of relying on spreadsheets and raw data, HarvestReplay helps a grower see things differently, Mantle says.

“Where it’s not just reliant on memory alone,” he says.

The audio briefings, Mantle says, help set a grower’s agenda for the day.

“This is like a game film review for Friday,” he says, adding that most HarvestReplay users listen to the briefing to start the day and then check in and see the HarvestReplay data in real time.

“This audio intelligence briefing shows what happened yesterday and what could have gone differently, then that turns into today’s playbook,” he says. “Then [growers] can also log in and check what’s happening in the moment on that day.”

Mantle says that information gleaned before the day begins can have a big impact on the day’s productivity.

“We see actually trends up hour to hour to hour to hour,” he says. “So, getting it right, right out the door where your people are knocking out of the ballpark, that’s where your bang for the buck is.”

The audio playbook is interactive, he says, where growers can push a button to ask for the top things to know about that day or for more context.

Rollout And Pricing

As of now, HarvestReplay is for current Innov8.ag customers growing blueberries, cherries and apples, Mantle says. He says the company has the most confidence in the data already in its system, but plans a broader rollout in 2027.

Pricing is at scale, he adds.

Return on investment, Mantle says, has a high potential, especially as HarvestReplay informs more efficiencies on the farm.

“When your farmers are running big crews, say 1,000 or more, even a few points of labor productivity has this massive impact straight to their profitability,” he says. “A 7% to 8% improvement in a 1,000-worker operation, doing something like just optimizing harvest station placements so workers spend less time walking, with scale it can actually translate into a seven-figure savings.”

Innov8.ag estimates HarvestReplay can offer an estimated savings of:

$25,000 to $100,000 for small-scale farms (less than 100 workers).
$100,000 to $250,000 for medium-scale operations (100 to 999 workers).
$250,000 to $750,000 and up for large-scale agribusinesses (1,000-plus workers).

Ask These Questions Before Taking the Plunge on New Precision Ag Tech

Tue, 06 Jan 2026 22:52:11 GMT

Editor’s note: This is the fourth of a series on questions growers should ask before investing in new ag tech. Because fresh produce is a high-value segment of agriculture, there are a lot of options available to spend money on, but asking the right questions before a purchase can save time, money and headache.

A lot of tech out there today either runs on data, generates data or both. Whether this is a connected internet-of-things-style irrigation sensor or a logging program to keep track of driver miles or a system that tracks harvest records that ties to payroll, there are some questions you should ask if data is involved in a new ag tech purchase.

They include:

Do I have what I need for this tech to be useful to me?
How usable (and actionable) will the data be?
How can I use that data with artificial intelligence (AI)?
Is this tech (and its supplier) right for me?
Does it work with my business’ workflow?

No. 1: Do I have what I need for this tech?

This one might seem a bit basic, but it still bears asking. You can’t really have a useful “internet of things” system if, for example, you can’t get an internet connection out to your orchard.

Lack of connectivity was a reoccurring issue that Liz Turner, marketing and special project coordinator for Croptracker, a precision ag tech software provider serving the fresh produce industry, encountered in her work early on.

“People were not taking into account that a lot of the data capture needed to happen where there wasn’t a lot of signal. And so missing data became an issue,” she says. That pushed Croptracker to develop offline data collection strategies. So, if your answer to “Do I have sufficient connectivity for this system to work?” is no, then you should likely ask the ag tech supplier if they have a system that can handle offline data collection.

But checking in on if you have what is needed for a data-focused ag tech system or solution can be less direct too. Because it takes a lot of data to help get the most out of data-focused ag tech, Chris Higgins, general manager of Hort Americas, a components and materials supplier for CEA growers, stresses the importance of having historical data well organized.

“Five years ago, I would have said that the largest growers were not even ready to have the tech conversation because their data was unorganized,” he says. The large growers have gotten organized since then, he adds, but smaller growers are still struggling with data.

“The smaller growers are battling the fact that, ‘Oh yeah, one guy knows all of this data.’ Now getting that guy to sit down and put it into an Excel spreadsheet is very, very challenging,” Higgins says.

Turner specifically recommends growers have an eye to the future of data — i.e. AI — when they think about the organization of their historical data.

“Thinking about ways that you’re organizing and maintaining your historical records, even without a big clear vision about how AI is going to replace XYZ processes in your operation, is really important because that historic data can become really valuable for future AI uses,” Turner says.

She urges growers to think about how their historic records and data, things like production practices and fertilization records versus yield records, might fuel a personalized predictive model in the future.

“It’s important to consider that you’re not just capturing [that data] for now or for a government regulation,” she continues. “You can be capturing it for the future. That historic data that’s specific to your farm and your location and your varieties is very valuable to you and should be treated as such.”

No. 2: How usable will the data be?

This was the most common recommendation from all sources The Packer talked to when it comes to investing in a new precision ag system or data-generating sensor. It’s a key question because, if the data coming out of the system isn’t usable to you, there’s little point in investing in it.

Unusable data will just become an expensive dust (or storage space) collector, as Steve Mantle, CEO and founder of Innov8.ag, a data- and AI-focused tech consulting group that works with permanent crop produce growers, experienced.

“There was one grower I met with recently who had a stack of basically two reams of paper soil fertility reports, and you could tell they’ve been sitting there for several months,” he says. The grower told Mantle he’d “paged through” the data a bit, “but, it was clear that it wasn’t really actionable for them. It stopped at the owner’s desk. It hadn’t trickled down to those other tiers that actually could make a difference with that data.”

Turner also highlights the importance of the data being usable to the right people in your organization. This involves thinking about who will be using the data and what format that data should take to be most useful to them.

“For the most part, people at a desk making decisions about packing or quality or sales aren’t going to be able to look through every single QC form,” she says. “So it’s thinking about not just ‘I need all of this information,’ but ‘I need my sales desk to be able to look at a single screen and read that information in a way that can actually action on it.’”

Actionability, the correct next step, is the key difference between useful tool generating usable data and a cluttered dashboard according to Roy Levinson, commercial lead for digital farming and water meters at Netafim North America.

“Remember that information is great, but if you don’t have the tools to actually execute on the information you are gathering, you are wasting resources,” he explains. “Be sure that if you are implementing a tool that gathers information, you then will also have the ability to apply changes because there is nothing worse than not having the ability to do anything when you know you should.”

No. 3: What can the data do with the help of AI?

Mantle recommends growers ask AI-related questions of their suppliers when it comes to selecting a new data-generating ag tech product or system.

For example, he suggests growers ask how the data, once extracted, could be used and translated by non-proprietary AI tools.

“I think a really good way to ask tech providers is: ‘How can I use other tools like large language models that are relatively accessible to people now to input and think about this data differently?’” he suggests.

“If the answer is you can’t or haven’t thought about it or ‘Let’s have a discussion about it,’ then maybe you’re not talking to the to the right guys,” he adds.

No. 4: Is this tech and its supplier the right one for me?

Mark Lukenbill, commercial leader at MileMaker, a trucking-focused software solutions arm of Rand McNally, also suggests growers ask some specific questions about the suppliers of any data-focused ag tech under consideration. For example, what is that supplier’s security history?

“Security is important,” he stresses, adding that growers should ask about data breaches and potentially steer clear of companies that have had them.

When it comes to the specific sensor system or program or other precision ag item you are considering, Lukenbill also urges growers to ask: “Will this put me into tech debt?”

“For instance, if I go buy an iPhone 11 right now, I’m in tech debt because even though it’s a new phone to me, that phone’s going to be obsolete in two years,” he explains.

The risks of going into tech debt by investing in older technology include the limitations of that older technology compared to competitors, plus needing to upgrade more often and sooner than otherwise, Lukenbill adds.

“Implementing especially big, big tech inside your business is a big undertaking,” he continues. “It’s going to create a lot of friction. There’s costs. The fewer times that you can do it over X period of time, the better off you’re going to be.”

No. 5: Does the tech work with my (team’s) flow?

Speaking of the potential of causing friction in your business or on your team, Turner recommends growers ask if a system or new data-generating tech item works with how things already work on your operation. She pointed to wisdom learned in supporting berry growers with harvest tracking and payroll solutions.

“[Berry pickers’] pay depends on them moving quick, so delays where you are making them wait and stop so that you can take account of that inventory is detrimental to everyone’s productivity,” she notes. “Is your software able to support that speed or does it slow it down?”

She gave the example of an inventory system that allows for quick picker badge scan followed by scanning in of the berry totes compared to a system that requires someone to type even one number into a device. While the former allows everyone to move in their rapid flow, the potential slowdown of the latter could mean it’s more efficient for everyone “to count everything at the end while the workers are doing something else,” she explains.

“Whereas the benefit of having it in real time means that traceability is more secure, it’s more one-to-one, you’re less likely to make mistakes,” Turner says. “But if that process is too slow, then it’s not worth it.”

Basically, if a potential new system or device doesn’t work with the existing workflow, it might be skipped, bypassed or otherwise not used to its fullest potential. Which means the team or people who will have to use it need to be consulted, she says.

“The biggest things that we try to get people to ask is ‘Who should be involved in making the decision?’” Turner says.

This usually translates to ‘Who’s going to be the most impacted?’ she adds. “Our most successful onboarding for payroll processes and inventory tracking and things like that happen when the person who’s going to use that data for reporting and payroll is in conversation with the person who’s on the field capturing that data.”

This usually means bringing in, at least, the field supervisors into the decision soon “so that they can test out the flows,” Turner says.

Next week’s installment of the Tech Questions series will dive into this topic even more by focusing on what questions growers should ask their team before investing in new ag tech. Catch the rest of the series here:

6 Ways Retail Grocers Can Prepare for the Rise of AI-Powered Web Browsers

Wed, 26 Nov 2025 20:06:39 GMT

As artificial intelligence-powered web browsers and agentic search tools such as ChatGPT, Perplexity and Gemini-enhanced Chrome become the default entry point for online discovery, grocery retailers are entering a new era of digital visibility.

For produce departments in particular, the shift is both an opportunity and a wake-up call: These AI agents will increasingly determine which retailers get recommended, which products surface first and which stores are seamlessly added to a customer’s cart.

Nate Barad, vice president of product marketing for Algolia , says this new landscape requires grocers to treat AI systems as a new type of customer, one that interprets product data, evaluates who has the best answer and then makes recommendations on behalf of human shoppers.

“Your fastest-growing user is the agent,” Barad says. “We have to prepare for an agentic customer the same way as a human customer. It’s not either-or.”

To stay visible in AI-driven search results — and to make sure fresh produce is recommended accurately — retailers must begin rethinking the way they structure, enrich and distribute their product data.

“Your fastest-growing user is the agent. We have to prepare for an agentic customer the same way as a human customer. It’s not either-or,” says Nate Barad, vice president of product marketing for Algolia.

(Photo courtesy of Algolia)

Barad recommends these six steps to help grocers prepare for AI-powered browsing:

1. Make Product Data Machine-Readable and Machine-Lovable

The first priority is product data quality, Barad says. AI systems need structured, consistent and complete data in order to understand and recommend products.

Barad breaks this into two categories:

Machine-readable: inventory data, nutritional information, naming consistency and the basic hygiene that allows AI tools to understand what the product “is.”
Machine-lovable: the kind of additional context that helps an AI agent “choose” your product over others, such as FAQs, conversational descriptions, storage tips, ripeness guidance and culinary applications.

This machine-lovable content is especially critical for fresh produce.

“Nutrition is going to be first,” Barad says. “But what the agent really wants is the information that isn’t normally in your index, like whether a chicken breast is used in chicken tikka masala. For produce, that could be what apples are best for a pie or which are better for keto.”

For grocers, this means enriching product catalogs with details they’ve never had to include before, such as:

Best uses (e.g., fuji apples for snacking; granny smith for pies).
Dietary attributes (low-sugar, keto-friendly produce items).
Ripeness stages and storage instructions.
Recipe compatibility.
Frequently asked questions.

This new metadata helps AI answer not just “Where can I buy Honeycrisp apples?” but also “Which apples should I buy for apple pie on Thursday?” — and recommend your store in the process, Barad says.

2. Prioritize Data Enrichment to Win the AI Auction

AI agents don’t simply crawl a website and show every retailer equally. They conduct what Barad calls a “mini auction,” analyzing multiple data sources before deciding which retailer gets recommended.

“Is it the type of information that gets you recommended, that gets you chosen as the answer by the agent?” he explains.

Conversational metadata, usage suggestions, recipe links and problem-solving content make a retailer more competitive in this silent AI bidding process.

For produce, Barad says this includes consumer-driven needs such as:

“What can I make with half a head of cabbage and two onions?”
“Which tomatoes are best for bruschetta?”
“What produce items fit a low-FODMAP diet?”

AI agents surface the retailer whose data best answers these questions, not necessarily the lowest price or the closest store.

3. Prepare for Zero-Click Shopping

As AI-driven experiences like Walmart’s GPT-powered shopping carts emerge, Barad expects browsing and purchasing to become more conversational, more personalized and less dependent on navigating retailer websites.

He warns retailers to prepare for zero-click environments, where a shopper may never visit their site at all.

“Your goals are going to be different,” he says. “You may not get a view or a click-through. Goals will be follow-up questions, voice add-to-carts and recurring personalized purchases: ‘Order me the same tomatoes I got last week; those were delicious. Never order me those tomatoes again.’”

This shift favors retailers whose product data is rich enough to support personalized guidance, especially for produce, which often relies on sensory qualities and contextual uses.

4. Be Ready for Problem-Solving Queries

AI-powered shopping is less task-driven (search apples, find results) and more solution-oriented (Which apple is best for my diabetic diet?). Barad expects grocers to see more problem-solving queries, particularly around:

Meal planning
Dietary restrictions
Substitutions
Nutrition comparisons
Cooking advice

This is where produce departments can shine. Shoppers will increasingly ask agents:

“What vegetables should I eat to boost fiber this week?”
“Which greens last longest in the fridge?”
“Suggest a produce-focused shopping list for a family of four.”

Retailers prepared with detailed, enriched data are more likely to appear in these high-intent recommendations.

5. Adopt the Model Context Protocol

One of Barad’s strongest recommendations is for retailers to ask their technical teams about Model Context Protocol, or MCP — a new standard that allows grocers to securely make their catalogs available directly to AI systems like OpenAI, Amazon and others.

“That’s the sanity check to the first question: Is my data even machine-readable? Can I transact in this new internet ecosystem?” he says.

For retailers, MCP is the difference between AI pulling unverified, scraped information from your website or AI using your official, structured product data, the version most likely to surface correctly in search.

6. Don’t Assume Your Current Data Practices Are Enough

Barad’s final caution to retailers: If you do nothing, AI engines will still pull your data, but in messy, incomplete ways that leave you at a disadvantage.

“If you’re doing nothing, that’s what’s happening,” he says. “Your data is being scraped. You have to prioritize the data you’re feeding these engines.”

Thus, retailers need intentional data pipelines, not passive ones, Barad says.

AI-powered browsers represent a foundational shift in how consumers discover and shop for groceries, even fresh produce. To stay visible, grocers must begin treating AI tools as customers who need clear, enriched and comprehensive information to make decisions.

With the steps Barad outlines, produce departments can do more than adapt, they can lead. In an era when shoppers increasingly ask AI “What’s the best apple for my pie?” retailers who invest in rich, machine-lovable product stories will be the ones whose aisles appear at the top of the answer.

Yixin Zhu Joins Flashfood to Lead Retail AI Efforts

Tue, 04 Nov 2025 23:04:47 GMT

San Francisco, Calif.-based retail marketplace Flashfood announced Nov. 4 that Yixin Zhu has joined the company as vice president of engineering. With more than a decade of experience building high-performing engineering organizations, Zhu will oversee the future of Flashfood’s data program to lead the company’s AI efforts.

“Yixin’s leadership background at disruptive companies and depth of technical skills are a perfect match for where Flashfood is headed,” says Jordan Schenck, Flashfood CEO. “The AI-powered products we’re developing enable us to move upstream from what we solve today, to address the billion-dollar opportunity in grocery inventory. Yixin’s expertise in designing best-in-class technology to solve complex systems will help us transform the way retailers operate.”

Zhu joins Flashfood from Archer, where he served as senior director of product engineering, leading teams that built software products for the emerging air mobility industry. He operated with a data-first mindset to create innovative products at the intersection of technology’s biggest applications: Transportation, manufacturing, defense, energy and AI.

Prior to Archer, Zhu spent five years leading technical program management at Uber, leading a global team of over 500 engineers to build the marketplace technology for all of Uber’s core products, including Eats, Rider and Driver, the systems that powered Uber’s multi-vertical last-mile delivery ecosystem.

“Flashfood sits at a point in our retail food system that’s seen less innovation than any other part of the chain. As an engineer, this is a very exciting place to be,” Zhu says. “Not only do we have the opportunity to build a solution that feeds the world, but we will completely change the way retailers manage their shrink and inventory today. By enabling retailers to be much more data-centric and leverage all the latest advancements in machine learning and AI, we’ll be able to bring in more shoppers, more revenue and ultimately more value for Flashfood’s retail partners.”

AgZen, Corteva Team up on AI-Powered, Retrofit Sprayer Tech

Fri, 24 Oct 2025 16:10:42 GMT

AgZen announces an agreement with Corteva to further “explore the commercial potential” of AgZen’s AI-powered crop spraying optimization technology, RealCoverage.

The news comes on the heels of Corteva’s big announcement on Oct. 1 , detailing the crop protection multinational’s plan to split its crop protection and seeds businesses.

AgZen, a tech startup spun out of MIT, is making a name for itself by pioneering feedback optimization for spray applications — a new approach the company thinks has potential to improve farmer outcomes and reduce crop input costs.

(AgZen)

AgZen’s first product, RealCoverage, is a retrofit kit that can be bolted onto any sprayer to measure and optimize the number of drops of agrochemicals applied to crops. The system features a boom-mounted sensor that analyzes the coverage and quality of spray applications in real-time, displaying actionable data to a tablet mounted in the cab. Farmers can use the data to optimize the physical settings on spray rigs, both self-propelled and pull-behind, to increase coverage.

The startup says its system works by leveraging AI and cutting-edge computer vision, and customers have used RealCoverage to save 30% to 50% on input costs.

Farmer Feedback

(AgZen)

Northwest Indiana farmer Bryan Brost slapped a RealCoverage system onto his Hagie STS 16 high-clearance sprayer to use on his waxy corn and soybean crops. He says it has helped boost his spray program efficiency overall by reducing application rates while maintaining optimal coverage throughout his 12,000-acre operation.

“The payback came in the first year,” he tells Farm Journal via text message. “We have increased our acres [covered] per day with less hours on the machine, the operator and the nurse tanks supplying product [to the sprayer].”

Corey McIntosh set the technology loose across his 4,000 acre spread in Missouri Valley, Iowa. He is looking forward to using the data to improve his application efficiency across the board. He’s also letting his neighbors and local retailer in on the secret.

“I was getting a chem shuttle refilled at [the] co-op, these guys have always been complimentary of our weed control, I asked them: ‘What percentage of leaf surface area do you think you are covering with your sprayers?’ One of their best operators said he thought 50% coverage. The salesman next to him said it would definitely be more than 60%,” McIntosh says.

“They were shocked when I told them we were at 9% to 10%, but nobody has had ever had a way to quantify this before,” he adds. “We are really looking forward to making improvements.”

Since launching on the market in 2024, AgZen says it covered more than 970,000 commercial acres of application across the U.S. on row crops and specialty crops.

Your Next Read: Breakthrough Fungicide Delivers White Mold Disease Control in Key Crops

How AI is Revolutionizing Mushroom Harvesting

Fri, 10 Oct 2025 15:27:08 GMT

4AG Robotics CEO Sean O’Connor says he founded the autonomous harvesting company with mushrooms in mind because he saw an opportunity to address labor shortages in agriculture.

“We thought through where is the largest issues when it comes to access to labor, the mushroom industry was clear No. 1,” he says. “Largely just because of the way the crop growth goes. The crop doubles in size every day. So, it grows by 4% an hour. It grows in a fairly chaotic manner, and so you have to be able to stay on top of the beds when they’re in that harvesting cycle. And because of that, there’s no seasonality to mushroom harvesting.”

And the 365-day nature of mushroom cultivation presented a unique opportunity and challenge, something he says fascinated the 4AG Robotics team.

“It’s also the most difficult crop to automate harvesting for,” he says. “It’s more delicate than a strawberry, more delicate than a tomato. It bruises easily. It splits easily. How you pick the mushrooms has a massive impact on the yield that a farm can produce.”

O’Connor also says he saw an opportunity to use data to help mushroom growers better understand the best time to harvest the mushrooms to maximize revenue.

“If you pick a mushroom at four o’clock in the afternoon, but the right time to pick is four o’clock the next morning, that’s 12 hours more growth,” he says. “That’s 48% more growth or revenue that you’re able to derive from that mushroom if you’re able to pick it at the exact time it should be fixed.”

He says that while it wasn’t exactly quick to develop the autonomous solution, he says it works on Dutch track growing systems, with no need to modify production systems to make it work. And he says customers wanting to add more harvesters helped push the company to this next funding round.

“That was really the catalyst for this financing, is that we’ve got robots that work,” he says. “We’ve got customers that want a lot more of them. And now we need to figure out how do we scale our manufacturing and scale the business?”

Riding the Mushroom Waves

He says the company has been riding the mushroom wave of growing interest among consumers, seeking out new flavors and eating experiences.

“You have this wonderful crop that’s literally growing at an extraordinary rate, and people are being able to scoop up more of it,” he says. “I think for us, we always knew that it was going to be a stable crop, one that was going to be a part of people’s diets.”

But O’Connor acknowledges automating mushroom harvest, or really harvest in general, has been a challenge for agtech companies. He says he and the 4AG Robotics team believed that with a specific approach, they could provide a solution. And artificial intelligence (AI) has really helped bring more actionable information to growers to make better harvesting decisions, he adds.

“The industry itself has gotten us to a place where we can make decisions with variable information that we couldn’t do when we had to build software or use heuristic methodology to make decisions,” O’Connor says. “That’s really powerful when you’re talking about optimizing when to pick each specific object that’s growing at such a quick rate.”

Future Goals

And as for the future? O’Connor says he expects to see some big changes in the mushroom industry quickly. While many farms converted to the Dutch track growing system in the late 1980s, he sees a similar precipice in the mushroom industry, where farms will decide on whether or not automation is a fit in the future.

“If they’re right, and they commit, they’re going to have a cost to them of being able to produce mushrooms at 60%, 70%, 80% of their competition, because you have a crop that turns over every week into revenue,” he says. “You’re effectively going to be seeing the farms that automate, scale and consolidate the industry, and those who don’t see financial woes and end up having to either be consolidated or go out of business.”

This summer, 4AG Robotics’ Series B funding round closed at 40 million Canadian dollars. Cibus Capital, an investment firm focused on sustainable food and agriculture, was the co-lead for this round. O’Connor says this funding will help the company get its technology into the hands of more growers and also help existing customers add more harvesters.

While his technology might seem like it could work in different commodities, as of now, O’Connor says the 4AG Robotics team doesn’t plan to expand into other crops in the foreseeable future. He says 4AG Robotics’ focus is on maximizing revenue for mushroom growers.

“We’re already seeing this now on one of our farms that we’ve gone from being a labor replacement solution into a yield generating solution that we’re no longer just giving you the same amount of mushrooms, but without the labor costs, but we can provide an increase in your yield every week, because our robots can pick for 24 hours a day and pick a mushroom exactly and should be picked,” he says.

O’Connor sees AI playing a greater role in the future of the company, where a grower can understand the quality of the clusters and weight as well as looking to yield forecasting and communicating that to customers.

“It won’t take us long to be able to say at the start of the week, if we see what the bed is looking like, we can communicate to the farm what they have to sell to their customers in advance of actually being able to pick it, in advance and starting its process of expiration,” he says. “All these little things where we can just give a slightly better logistics engine to these farms to become the operating system allows them to be most efficient using the artificial intelligence that we create or adding more automation into the packing room of these farms as well, or other opportunities like that.”

Simbe for Merchants to Bring Real-Time Shelf Visibility to Produce and Beyond

Tue, 30 Sep 2025 12:20:43 GMT

LAS VEGAS — Simbe Robotics unveiled Simbe for Merchants at Groceryshop 2025, a new suite of tools designed to give merchandising teams unprecedented visibility into what’s happening on store shelves. The rollout marks Simbe’s first purpose-built solution for a specific retail team, reflecting the growing pressure retailers face to execute flawlessly in a hypercompetitive environment.

Caitlin Allen, senior vice president of marketing for Simbe Robotics, explained in an interview with The Packer that the launch helps solve for real-time visibility in high-volume, high-velocity categories like produce.

“Today, 60% of planograms have inaccuracies,” Allen says. “Our product reads the shelf and gives merchants the ability to look at what we call a ‘realogram’ — basically what the shelf actually looks like — instead of waiting days for field teams to send pictures. That’s not just efficiency; it’s sustainability, too, since it cuts out all those unnecessary car trips.”

“Retailers win or lose at the shelf, yet actual shelf visibility has always been limited to spot checks and occasional store visits,” says Tom Gehani, vice president of product for Simbe. “With Simbe for Merchants, teams have access to the daily insights they need to ensure the right products are where they belong, displays are set right and vendors are aligned on execution.”

Why It Matters Now

The timing underscores a critical industry challenge. Recent Coresight Research shows retailers lose 5.5% of sales and 5% of margin to in-store inefficiencies, while shoppers consistently cite empty shelves and pricing errors as their top frustrations, the company says. With fresh produce playing an outsized role in driving grocery trips and brand loyalty, the implications for produce departments are especially strong.

Allen notes that fresh is where these gains may be most visible.

“Fresh is a big draw for shoppers into traditional grocery stores in a world where online incumbents matter so much,” she says.

By combining robot-powered scans with fixed sensors, Simbe captures critical details several times a day, including whether produce is available, properly priced and stocked at peak demand.

“Let’s say bananas tend to run low around 5 p.m. Our system can flag that in real time so store teams can replenish before customers walk away disappointed,” Allen says.

“Retailers win or lose at the shelf, yet actual shelf visibility has always been limited to spot checks and occasional store visits,” Gehani says. “With Simbe for Merchants, merchandising teams have access to the daily insights they need to ensure the right products are where they belong, displays are set right and vendors are aligned on execution unlocking measurable sales and margin gains.”

Simbe Robotics unveiled Simbe for Merchants at GroceryShop 2025, a new suite of tools designed to give merchandising teams visibility into what’s happening on store shelves.

(Photo: Jill Dutton)

Expanding the Platform

Simbe also revealed Simbe Mobile 2.0, a redesigned app built with store associate feedback. The update promises streamlined navigation, smarter task management and instant store health snapshots. It will be available in app stores next month.

For produce leaders, the launch signals a sharper set of tools to fight margin erosion and shopper frustration. With perishables among the highest-turnover items — and central to how grocers differentiate themselves — timely execution can mean the difference between loyalty and lost customers. By digitizing the shelf, Simbe aims to make produce execution not just measurable, but predictable.

“Solutions like Simbe’s can’t just live with store operations,” says Sean Spillane, former senior vice president of strategy and real estate for Stop & Shop and a strategic adviser for Simbe. “Reflecting on my experience, I recommend bringing merchants into the shelf digitization program early. They have so much to gain from this data, and the value is undeniable.”

The announcement of these new apabilities reflects the growing maturity of the shelf digitization space, Simbe says. Building on milestones such as Simbe’s Strategic Advisory Board and advanced capabilities for fresh departments, Simbe for Merchants leads the way as the first in a series of tailored, team-specific solutions designed to extend store intelligence across the retail organization as foundational data infrastructure that powers modern artificial intelligence initiatives, the company says.

Artificial Intelligence Joins The Fight Against Weeds, Insects And Disease

Fri, 08 Aug 2025 13:53:42 GMT

The crop protection industry needs a reboot, according to Tony Klemm.

As CEO of Enko, a crop-protection startup, he says the company is taking a different approach to solving one of agriculture’s biggest problems – developing safe, effective and sustainable crop protection products that can be brought to the marketplace faster and more economically.

Traditional discovery pipelines for herbicides, fungicides, insecticides are not keeping pace with real challenges farmers face, such as resistance issues, he told Chip Flory, host of AgriTalk on Thursday.

A 2024 study funded by Crop Life International reports the costs associated with bringing a new active ingredient to major U.S. and European markets now top $300 million. In addition, the survey says the average lead time between the first synthesis of a new crop protection molecule and its subsequent commercial introduction is now over 12 years.

Part of the long development time required is related to regulatory hurdles. “There’s just increasing demand for meeting environmental safety needs, rightfully so,” Klemm says.

A Paradigm Shift
Enko, based in Mystic, Conn., is using artificial intelligence (AI) and a machine learning discovery platform to guide the company’s research and development efforts. Klemm describes the strategy as a paradigm shift from the current industry practices for how small molecule crop protection discovery has been done.

“We use DNA-encoded libraries, and these libraries allow our scientists to explore this massive, diverse chemical space in a very targeted, automated and expansive way,” he says.

The technology allows Enko scientists to look at billions of molecules and screen them for safety and efficacy and, in the process, develop them faster and more economically.

“We still have to take the regulatory journey that, right now, no one’s figured out a way to expedite,” he notes. “But getting to that regulatory queue faster and better on the front side is really what’s bringing us that cost savings, that efficacy and is going to allow for more products to be put into the regulatory queue in a faster manner.”

Progress To Date
So far, Klemm says Enko has delivered about 50 active programs that cover all facets of weeds, insects and disease. Many use novel or new modes of action that Klemm believes will help farmers fight resistance issues, such as herbicide resistance in Palmer amaranth and pigweed.

“We’re really working on how we can bring new modes of action to farmers, give them fresh tools to win that fight. And our chemistries work using fewer active ingredients, from perspective of the load on the acre, so we’re designing safer chemistry for the future,” he says.

Specifically, Klemm says Enko recently announced a new grass herbicide is in the pipeline for the European cereals market for control of black grass. The company also has conducted field trials for corn and soybean products in the U.S. that he anticipates are five to 10 years away from market launch, depending on how long they take to move through regulatory channels.

Your next read: Bayer Affirms Support of Glyphosate, Optimistic for a Future with Over the Top Dicamba Labels

University of Florida Research Looks at Future of AI in Sprays

Sun, 27 Jul 2025 20:20:07 GMT

Nathan Boyd, a University of Florida Institute of Food and Agricultural Sciences (IFAS) professor of horticultural sciences, says about 80% to 90% of his research now centers around artificial intelligence (AI). While precision ag is nothing new, he says how AI will help integrate into precision agriculture shows so much promise.

“If you go back and you can almost hear in the articles this desire by the authors that they wish they could do certain things, but they did not have the tools to do it,” he says. “And AI is what, all of a sudden, gave us the capability to do all these things that they had wanted to do for decades but couldn’t.”

Boyd says he’s paired his field trial expertise with Arnold Schumann, a IFAS professor of soil, water and ecosystem sciences, to unlock the future of agriculture.

“I think it’s the next agricultural revolution,” he says, “I think it’s going to completely transform how we grow food. But I think, we are only at the very beginning. We have not yet seen what this is going to be.”

In Boyd’s recently published findings, a research team compared the use of AI technology with herbicide applications on weeds in tomatoes in three locations. The research team compared results of herbicide applications on weeds growing in bare soil between beds covered with plastic mulch; herbicide applications in transparent holes on top of the plastic mulch beds; and targeted herbicide applications on nutsedge, which punctures plastics, with the use of machine vision to locate the weeds.

The results, Boyd says, show targeted spray methods can work effectively for vegetable crops and lower input costs related to weed management by 75%. He says there’s also ongoing projects in strawberries, tomatoes, peppers, watermelon and more.

But part of the motivation to study tomatoes is that there are herbicides registered in tomatoes to treat nutsedge. One thing Boyd says is that the industry needs to understand that AI is a tool that works in tandem with herbicides that target weeds.

“AI is not going to solve everything,” he says. “It’s a way to make things more efficient, but it doesn’t remove the need.”

Boyd says his research team has had the support Artificial Intelligence Academic Initiative Center in its research and has worked with growers to develop the techniques.

“We said, ‘OK, we need to talk to growers from the very beginning and then so that when we design it, it’s actually designed in a way that they could use it,’” he says.

The other thing Boyd says was important with his study was to show that the AI technology can integrate into a grower’s existing system without having to modify growing systems.

“We wanted to design it in a way that could be used in current systems,” he says. “We’re also trying to design it so it’s adaptable. We have one basic AI program and system that we’re designing to use across multiple crops.”

Boyd says growers also want to be able to maintain their equipment.

“We have to design it as a modular approach,” he says. “So, you could kind of plug and play, so you can remove things and replace them if they need to be fixed.”

Boyd also says growers could see benefits beyond just a reduction in inputs. Targeted sprays can reduce the chance for crop damage from herbicides. Plus, AI systems with vision and mapping techniques can help growers better monitor what’s going in in their fields. And this goes beyond just scouting and treating weeds; AI systems will also help growers catch the presence of diseases earlier.

“Most scouts look at less than 1% of the plants in the field, just because of time restraints, but this machine going through would be looking at every single one,” he says. “That’s a big difference and much earlier detection means not only do you save because you’re not only spraying where it’s needed, but you also save because you’re able to spray things earlier.”

How AI Brings Intelligent Solutions to the Produce Aisle

Wed, 23 Jul 2025 20:49:30 GMT

Retailers are increasingly turning to artificial intelligence in the produce aisle to improve shopper experience, optimize operations and boost sales. Grace Schroeder, CEO of Slingr , shares how AI is already reshaping the fresh produce landscape — and where it’s headed next.

“We’re seeing a patchwork of point-of-sale, inventory and spreadsheet systems in grocery that don’t always talk to each other,” she says. AI has the potential to stitch those together, she adds, pulling actionable insights from all the disconnected data to improve everything from inventory management to marketing outreach.

Slingr is an AI-powered automation platform designed to help companies streamline operations, reduce waste and boost profits by connecting their systems, data and workflows into one intelligent, customizable solution. Rather than selling a single software package, Schroeder says, Slingr works with grocers to assess their existing systems and build custom automation solutions. These solutions often start with identifying areas where employees spend time on repetitive, manual tasks — like tracking produce sales, checking equipment performance or pricing inventory — and developing intelligent tools to streamline those processes.

For example, if a cooler frequently fails and causes produce loss, AI can recognize patterns in sensor data and trigger predictive maintenance before spoilage happens, Schroeder says. It can also monitor how much produce is sold at full price versus discount or clearance, helping grocers refine pricing and reduce waste.

Slingr CEO Grace Schroeder

(Photo courtesy of Slingr)

But AI’s value goes beyond the back end. Schroeder says smart automation can also enhance customer service text or email responses to inquiries or even chatbots that understand typical questions and generate human-like answers.

“There’s a lot of fantastic AI-powered marketing tools out there,” she says, adding that automation helps retailers stay connected with shoppers in a more timely and personalized way.

Still, adoption won’t be without its challenges. Grocers who embrace AI will gain a cost and margin advantage, Schroeder says, while those who delay may struggle to compete.

“You used to be able to get by with spreadsheets — even million-line ones — but that’s no longer sustainable,” she says. “The industry needs to move toward systems that can communicate and interpret data in real time.”

What’s promising is that this digital shift doesn’t have to be overwhelming.

“You don’t have to automate everything all at once,” Schroeder says. “Start with the areas where you’re losing money or spending too much time. From there, you can build.”

Ultimately, Schroeder says she sees a future where AI enhances, instead of replaces, the human expertise that makes grocery operations successful.

“It’s evolutionary,” she says. “Now, the automation we can bring is a lot more elegant, it’s easier and it’s a lot more human.”

How Social Media, Sustainability and Smart Tech are Reshaping the Industry

Thu, 19 Jun 2025 18:19:02 GMT

Technology is transforming the fresh produce supply chain from farm to table. Here we explore how social media, sustainability and smart tech are reshaping the industry.

Social Media: An Unlikely Driver of Fresh Produce Changes

Some technology is less tangible than an optical sorter, but that doesn’t change the impact it can have on the industry. Consider social media. While it is not growing technology in a classic sense, it is a vehicle that has driven on-farm activities as much as any farm truck.

For example, in the world of sweetpotatoes, the purple-skinned, white-fleshed murasaki varietal is seeing increasing consumer demand to the point that many farms are planting more and more acres around it and offering new product lines. Why do consumers want it?

“People shop with their eyes, and they eat with their eyes,” said Autumn Campbell, sales manager for Arkansas sweetpotato grower Matthews Ridgeview Farms. “[Murasaki sweetpotatoes] are beautiful, so they look good on plates and combined with other sweetpotatoes.”

That visual appeal works well in the world of social media, according to Michelle Grainger, executive director of the North Carolina Sweetpotato Commission.

“The culinary world is always looking to do something different on the plate and to have pop,” Grainger explained. “And that ties into your social media influencers who are taking the pictures and are posting about the latest creation that they were able to enjoy.”

From Social to Sustainability

Since social media means a shorter distance between farm and fork, it allows consumers to gain information — and demand greater action — on topics
that are important to them. Topics that include sustainability.

Sustainable packaging is an area of growing interest. In the past three years of The Packer’s Fresh Trends reports, respondents have consistently rated biodegradable packaging as a key way the fresh produce industry can promote sustainability.

“Consumers are really interested in packaging right now, and for good reason; it’s the biggest and usually first touch point most consumers are going to have with their fresh produce,” said Jeana Cadby, environment and climate director for Western Growers Association, which represents fresh produce farmers in Arizona, California, Colorado and New Mexico.

She pointed out that there are a lot of ways packaging can be sustainable. There is the direct way that most consumers likely think about — packaging material that is recyclable or biodegradable, or as minimal as can still get the job done — but she said that roughly 90% of the work of packaging is done before a product reaches the consumer.

Take lettuce, for example: “As soon as you pick it, the job of the packaging is to keep it as alive and fresh for as long as it possibly can. That itself is a huge technological marvel,” she said. “When we think about how our produce network is set up, it’s actually really incredible how far a bag of lettuce can travel and still be fresh when it comes to consumers.”

The more bags of lettuce — or clamshells of strawberries or packs of mushrooms and so on — that reach consumers fresh and stay fresh long enough to be eaten means less food waste, another angle of sustainability.

Smart Tech Has Changed Labor

Victoria Lopez — marketing and business development manager for Fox Packaging and Fox Solutions, fresh produce packaging makers and equipment distributors — sees technology in produce primarily answering questions of sustainability.

“How can we be sustainable, not just with packaging materials, but how can we be responsible with resources?” she said. “How can we be responsible with talent?”

Large infrastructure using artificial intelligence (AI) is helping with that now, Lopez explained.

“The advancements of technology have supported the flexibility and adaptability of our ability to process and get food to the store shelves,” she said. “With technologies, we’re able to determine what size produce is and how fancy it is.”

She gave the example of the Newtec Celox-P-DUAL-UHD, an optical sorter distributed by Fox Packaging that uses AI to assess photos taken of every potato that goes through for characteristics such as size and flaws across several different potato varieties.

“It can sort up to 13 different categories, identifying decay, rot, greening, holes,” Lopez described, adding that the programming allows it to adapt to new defects as it encounters them.

“The AI has been trained to recognize good versus bad potatoes just like how a human would learn, but faster,” she added.

Though the point of smart technologies isn’t to replace human labor, when it does, it also opens up the opportunity for operations to reallocate their talent, Lopez continued.

“It creates more growth opportunities for your workforce, whether it means becoming a lead, a supervisor, perhaps even a manager, or creating new departments — wherever there may be a gap,” she said, adding that identifying gaps is the point of automation. “Where is too much time being consumed in the process, and how can we make it more efficient? That’s where the innovation and creativity comes into place with a lot of this technology.”

Ultimately, she called AI a powerhouse for the future of the fresh produce industry that will arm people like operations and facilities managers to get deeper insights with real-time information.

Focus on a More Efficient Future

AI in agtech isn’t new, despite the growing public attention and interest.

“We’ve had laser weeders that use machine learning to identify weeds and help with reducing a lot of that labor that goes into weeding a field,” Cadby gave as an example. But she thinks the industry has only just started scratching the surface of the possibilities.

“It’s really fun to talk about the really big, exciting things like the giant laser weeders or the handpickers that are going to come out and increase efficiency and pick in the middle of the night,” she said. The big advancements will come in the “boring stuff,” she continued, technologies that answer the question, “How do we just make things more efficient?”

“For example, you have the laser weeder that’s out there weeding, but maybe it can also be taking images and tell you when you it’s identifying some sort of pathogen in the field,” she said. “Maybe it can do an early, super targeted identification of that pathogen, and it helps to capture that before it spreads to the rest of the field. That’s an example of doing more using the tools that we already have, or fewer passes in the field or resources generally.”

Less-flashy tools that make it easier for producers to make good, data-based decisions are also where Cadby sees the future of technology in produce. She offered Western Growers’ Green Link project as an example of such an effort that examines how to advance food safety with innovative technology.

“It’s using aggregated data to pull the industry together and design this more resilient and safer ecosystem for growers. I think that’s really cool,” she said. “It’s just exciting that we have these kinds of tools at our fingertips now that we can use to really make these educated decisions.”

Lopez described the future of technology in the fresh produce industry as being “turnkey;” smart technologies and partners that help operators throughout the supply chain make informed decisions from data that translate into improvements.

“I think technology is broader than people give it credit for,” she said, pointing out that it’s not just machines and software but also new approaches and processes. “Technology is hand in hand with innovation, and innovation is a supporter of anything that helps make improvements.”

Your next read:
Technology Goes Hand in Hand With Innovation in Produce

John Deere-Sentera Tie Up: Here’s What We Know So Far

Tue, 27 May 2025 15:40:37 GMT

John Deere has announced its acquisition of Minnesota-based aerial optics innovator Sentera. Although specific details are few and far between this early in the process, here’s what we know so far:

The two companies have a long history. John Deere was the first enterprise customer Sentera signed onto its system over a decade ago, and the two companies have had an API link in place between Sentera’s drone management software and John Deere’s Operations Center since 2016.
Financial details are not being disclosed. We do know the deal is not subject to any further regulatory or shareholder approvals.
In a similar fashion to the Blue River Technologies and Bear Flag Robotics acquisitions, Sentera will maintain its independence as a free-standing business unit. Once fully integrated into the Deere family, Sentera will operate under the John Deere Intelligent Solutions Group (ISG) framework. Sentera leadership will remain at its St. Paul, Minn., headquarters.
For the time being, no major changes are planned for either company as we head into the heart of the summer crop scouting and spraying season. The two companies anticipate having more details to share about the nuts and bolts of the acquisition this fall.
The two groups are a natural fit. Sentera is aggressively marketing its SmartScripts drone weed mapping program, and the technology is complimentary to John Deere’s Operations Center and its See & Spray and ExactApply application technologies. One driving force behind this deal, Farm Journal is told, is Deere’s motivation to integrate more real-time agronomic data into its Operations Center platform, and Sentera’s aerial data capture capabilities can help make that happen.

(John Deere)

A deal to lift both boats. John Deere has built up a deep bench of artificial intelligence, machine learning and autonomous technology expertise within ISG, and Sentera has a long track record of aerial sensing and camera payload innovation. Considering how many cameras and sensors are included from the factory on new John Deere machines and within its Precision Upgrades retrofit kits, there should be a healthy cross pollination of sensor and camera innovation between Urbandale, Iowa, (where ISG is based) and St. Paul, Minn.
Sentera can help make See & Spray even better. SmartScripts uses drone-based imaging to scan a field and build a weed pressure map which is then loaded onto the sprayer’s in-cab computer. Now the sprayer operator can see exactly where weeds are in the field and focus their spraying efforts there first. There’s also a logistical and planning aspect to SmartScripts: by knowing exactly how many weeds are present in the field, and even what type of weeds are there, an adept operator can have the right active ingredients premixed and the exact amount needed loaded into the tank or staged nearby in a tender truck to keep that sprayer running all day long.

“Farming is becoming a very sensor and data-centric business, and in our opinion, there isn’t anyone doing it at broad scale today better than John Deere,” says Eric Taipale, chief technology officer, Sentera. “The way we can bring these data-driven insights and improve grower outcomes — it’s just what we’ve always been about. It’s what John Deere is all about. There’s such a great mesh between the two cultures, the objectives and the mission of the two organizations.”

Joseph Liefer, global technology marketing lead at John Deere, adds, “We’re excited about how this complements our existing portfolio with See & Spray, and then not just that (product). Now a farmer with an individual nozzle-controlled sprayer from any manufacturer can also leverage this technology. A drone can fly their field, generate a weed map, turn it into a prescription in Operations Center and the machine can go execute the plan. From an ag retailer standpoint, that might have a mixed fleet, and this gives them more tools in the toolbox to do targeted application for growers and help them save on herbicide. We view this deal as complementary to our overall tech strategy.”

Your Next Read: MAHA Report’s Surprising Stance on Glyphosate, Atrazine Explained

Is ‘farming without farmers’ the future of ag?

Thu, 15 May 2025 18:53:50 GMT

The rapid evolution of technology, robotics and artificial intelligence has brought continuous and dramatic change to every industry — including, and perhaps particularly, agriculture. To discuss the growing role of AI across agriculture and the fresh produce industry, The Packer sat down with Rizwan Butt, vice president of product management for iTradeNetwork, in a fireside chat webinar on May 14.

“Imagine a future where we are literally farming without farmers,” Butt said. “And that may sound scary and anxiety-inducing for a lot of people, but … the question we have to ask ourselves is: Are we of those who are going to embrace these realities and adjust and adapt to them, or are we going to be apprehensive about them and try to avoid them for as long as possible?”

Agriculture is already employing robotics to aid in everything from planting to crop protection to harvesting to selecting for quality and much more, Butt said.

“This is not some scientific sci-fi future. This is a reality today that’s being done either in labs or in experiments or actually being deployed in the real world,” he said. “So, there is going to be a point in time where we have to accept the fact that there are going to be machines and AI tools that are going to augment what we do as human beings and make that job much more productive, much safer and much more optimized.”

As demands on the global food supply are expected to increase by 40% to 70%, while at the same time the labor pool continues to shrink, the need for AI in agriculture is real and the time is now.

For more on how AI is helping agricultural supply chains evolve from reactive to predictive as well as how retailers and growers can now anticipate disruptions — whether from weather, demand shifts or logistics breakdowns — and adapt proactively, watch the webinar now.

Watch on demand: Register here to download and watch the webinar

Procurant debuts AI solution

Mon, 14 Apr 2025 16:11:18 GMT

Procurant, a provider of technology solutions for the food industry, launched its Procurant Fresh AI, an artificial intelligence solution designed specifically for perishables category management in retail grocery. Procurant said in a news release that Procurant Fresh AI leverages its extensive expertise in the highly volatile perishables category to reduce costs, streamline decision-making, eliminate complex reporting and deliver accurate insights to executives and front line users alike.

The company said Procurant Fresh AI helps retailers maintain a competitive edge by providing accurate insights within an AI-powered interface that streamlines multiple steps and processes associated with category management. Procurant said its AI solution first sorts through volatile and often unstructured information to organize data into a secure, AI-ready format. Then, in plain language, users can ask questions about the category and act on new insights.

Procurant said Procurant Fresh AI can handle questions of varying complexity and instantly delivers both narrative and graphical results. Simple questions such as, “How many cases of apples did we purchase in week 32?” can evolve into more complex questions such as, “Which distribution center paid the most for Fuji apples in Q1 of 2023?”

Key features of Procurant Fresh AI include:

Real-time insights with instant visibility into category performance, inventory levels and supply chain trends.
Stand-alone or integrated solutions with data from existing internal systems, or as a component of the Procurant Commerce procurement platform.
Combines data from multiple systems into one easy-to-use interface, simplifying operations.
Offers actionable recommendations for pricing, promotions and inventory management to optimize performance.

“At Procurant, we’ve built our reputation on helping customers tackle the unique challenges of the perishables category,” said Eric Peters, founder and CEO of Procurant. “Procurant Fresh AI takes that a step further by providing retailers with the tools they need to access real-time insights and to make informed decisions in a market where timing and precision are critical. They now get direct, precise answers to their most pressing questions — no more sifting through endless reports or juggling disconnected systems.”

Carbon Robotics adds autonomous tractor solution

Mon, 31 Mar 2025 18:38:41 GMT

In a move to help growers maximize equipment and address labor shortages, Carbon Robotics launched its Carbon AutoTractor, an autonomous solution installed on existing tractors.

Carbon Robotics’ founder and CEO Paul Mikesell says its Carbon AI will power remotely monitored tractors to help specialty crop growers deploy laser weeders for almost around-the-clock production.

“With LaserWeeder, farmers want to run them as much as they possibly can, but it’s hard to find labor,” he told The Packer. “It’s really hard to find labor to do the tractor driving. It’s hard to find labor to do these late midnight shifts. It’s hard to find people to do all the different tasks you want to do with the tractors.”

The Carbon AutoTractor features two core components: the Carbon Autonomy Kit and the Remote Operations Control Center. Mikesell said operators in ROCC handle any obstructions through monitored autonomy and take over the autonomy system, so production continues. He said growers, then, don’t have to worry whether an autonomous task gets completed.

“They’re trained in using Carbon Auto Tractor,” he said. “They know how to do the functions that the farmer wants to do in the field. And then, whenever there’s something that comes up, they can literally change drive the tractor remotely, and get through whatever obstacle it is, and then keep moving.”

Mikesell said the Carbon Auto Tractor will currently work for LaserWeeder tasks, ground prep such as mulching, mowing, discing and more, but there are plans to expand its capabilities in the future.

The Carbon Autonomy Kit is initially compatible with John Deere 6R and 8R Series tractors, requiring no permanent modifications and installation completed in less than 24 hours. Once installed, tractors can toggle between autonomous and manual operation as needed.

“It bolts on and then that you plug into the harness in the inside on the inside cab, and there’s a box that mounts on the window that you can turn it on and off,” Mikesell said.

The Carbon AutoTractor system includes RTK-accurate GPS, 360-degree cameras and radar-based safety sensors, as well asphysical, remote and mobile e-stops connected via a high-speed, low-latency satellite link.

“We have better visibility from the cameras on the roof than you do from the inside the cabin,” Mikesell said.

And for those growers who might be reluctant to go to an autonomous tractor, Mikesell said the Carbon AutoTractor is designed to help growers better deploy farm labor where it’s needed most.

“You want somebody there to inspect or notice problems with your irrigation or things of that nature. You’ll still want to have those people around, but the point is that they don’t have to spend all that time driving up and down the rows to do the simple task,” he said. “They can then spend their time focusing on figuring out where or if there’s issues and how to address other problems and it relieves the constant need to be driving the tractor all the time.”

Mikesell said this solution also offers growers the option to deploy tractors at night for weeding or when the nighttime temperatures are cooler. This also helps growers maximize return on investment by being able to run the autonomous solution all the time, he said.

“We found that lot of people want to run their LaserWeeder 24/7 because they get a really good ROI or more crops they can put it under, but they just can’t find the operators to run it 24/7,” he told The Packer. “If you can run it, 24/7, you can double the hours in a typical season and you can get that tool doing everything you need it to.”

Carbon Robotics said the autonomous solution works seamlessly with its LaserWeeder, automatically adjusting speed to optimize weeding performance based on weed type, size and density, which can boost coverage by up to 20% compared to manually operated systems.

Brandon Munn, farm manager with Columbia Basin Onion, has worked with the Carbon Robotics team on this autonomous solution.

“With many of our tractors and LaserWeeders running autonomously with Carbon AutoTractor, we’re able to operate more hours, address labor challenges and make night shifts safer and more reliable,” Munn said in a news release. “This isn’t just automation; it’s a practical solution that’s fundamentally changing how we farm.”

Additional Details Come In On AutoTractor
Farm Journal talked with CEO Paul Mikesell to see what else we could learn about the system and what makes it different from other tractor autonomy kits on the market. Here’s a handful of bullet points breaking down what we uncovered:

Don’t Call It A Retrofit - Because the AutoTractor kit doesn’t effectively alter or change anything mechanically on the tractor itself, Mikesell says he prefers to refer to it as a “augmentation kit.”
Satellite Connectivity Changed The Game - When Mikesell and his team started this project back in 2023, connectivity was a limiting factor in enabling a tractor to safely operate with complete autonomy. That is no longer a limiting factor as developments in the stratosphere like SpaceX’s StarLink and Intellsat’s low earth orbit constellations have provided the necessary latency and bandwidth to make driver-less operation safe and viable.
Pricing Is Still TBD - Pressed on how much the system will cost from an up-front investment standpoint, Mikesell told us that “we’re still fine tuning that price.”
Expect An Hourly Fee - Mikesell did confirm that the technology will carry a per-hour fee. He says that fee will track closely with what the user would pay a local machine operator to run the tractor. That could mean a per-hour fee anywhere from $15 in the Midwest to upwards of $25 per hour in high-wage markets like California and Washington. “We’re trying to save you money by not having to worry about travel time out to the fields. There’s no lunch break. You don’t have to worry about paying overtime. This machine will do as many double shifts as you want, and we’re still employing people to do all the monitoring. So we have a very skilled and qualified group of people that are doing all the monitoring. So that’s kind of the model: we charge you per hour to run this machine for you and we’ll work with you on what jobs you want done and how you want it done and make sure that everything is handled appropriately.”
Remote Operators With Tractor Experience - Mikesell says his remote operators that task and oversee the driverless tractors for farmers get a crash course in how tractors are used on your typical farm. “Just being out there in the field long enough to understand the size of things that are around you and just kind of what a field looks like and how things are laid out, makes a huge difference when you’re trying to drive remotely,” says Mikesell. “Even though you have a better view driving remotely, because you have a nice 360 degree view off the roof, having some concept and understanding about the size of things and kind of what everything looks like helps quite a bit.”

Your Next Read: Using tech to target food safety threats

Instacart enhances app personalization with AI-powered Smart Shop

Tue, 18 Mar 2025 20:08:15 GMT

Instacart has introduced Smart Shop, which its says leverages generative artificial intelligence and machine learning to create a more personalized grocery shopping experience.

With Smart Shop, Instacart is making online grocery shopping more intuitive by analyzing customer habits and dietary preferences to surface the most relevant products faster, according to a news release.

In addition to Smart Shop, Instacart unveiled new AI-powered Health Tags, which provide detailed and transparent nutritional information across the catalog, and Inspiration Pages, curated destinations within the Instacart experience featuring expert-backed health recommendations and shoppable recipes, the release said.

The first Inspiration Page was developed in collaboration with the American Diabetes Association and features evidence-based nutrition guidance and diabetes-friendly grocery and recipe recommendations, Instacart said. Launched to coincide with National Nutrition Month, these offerings make it easier for consumers to discover relevant products based on their unique health and lifestyle preferences on Instacart, the company said.

“At Instacart, we want to turn the ordinary task of grocery shopping into a delightful, personalized shopping experience that takes the mental load out of finding the exact items that meet your preferences,” said Daniel Danker, chief product officer for Instacart. “By combining our new Smart Shop technology, Health Tags and Inspiration Pages, we’re not just improving online grocery shopping — we’re reimagining it, making it seamless to go from intention to action. By customizing your shopping journey to match your personal health goals or fit your dietary restrictions, we can unlock possibilities that weren’t even on the table before.”

Instacart’s latest features enable consumers to shop according to their unique dietary and household preferences, simplifying the process of finding relevant products and making more informed grocery choices, the release said.

Powered by Instacart’s catalog of 17 million unique items and its proprietary dataset of millions of grocery shopping journeys, Smart Shop leverages insights into nuanced consumer behaviors, the company says. Smart Shop technology applies behavior analysis and affinity modeling to analyze real-world shopping habits, identifying patterns in user preferences with precision. To further refine this understanding, large language models improve personalization by identifying the most valuable signals for predicting preferences to enhance accuracy.

For example, Instacart said inferring a customer’s preference for low-carb products is complex because many everyday items are naturally low in carbohydrates, making it difficult to determine true intent. Traditional machine learning models rely on large datasets and extensive feature engineering to differentiate between routine purchases and deliberate low-carb choices.

Items like almond flour or cauliflower rice strongly indicate a low-carb preference, whereas staples like chicken breast, though low in carbs, are less indicative of intent in this context, the release said. Large language modelssimplify the process by analyzing product categories and detecting intent-based patterns in user behavior. They pinpoint deliberate low-carb choices and filter out noise from irrelevant signals like purchases of naturally low-carb items.

By automating the identification of meaningful patterns, large language models eliminate the need for exhaustive tagging and manual feature engineering, drastically reducing the time and effort required to create highly accurate, scalable predictions, while also enabling more precise and personalized recommendations — even in scenarios it hasn’t been explicitly trained on, the release said.

As users shop, Smart Shop evolves its understanding in real-time, dynamically adjusting digital aisles to prioritize the most relevant products based on inferred preferences, while seamlessly adapting to changing behaviors over time, said Instacart. When confidence in a preference is low, Smart Shop proactively engages users with clarifying questions like, “Show more low-carb options?” These interactions, combined with Instacart’s proprietary shopping data, advanced machine learning models and large language models reasoning, continuously sharpen Smart Shop’s ability to predict user needs and refine recommendations. By combining structured insights with pre-trained AI, this advanced approach drives unparalleled precision and truly personalized grocery experiences, the release said.

Instacart also highlighted the following:

Customizable Smart Shop preferences — To further tailor the Smart Shop experience, customers can modify their shopping preferences, selecting from 14 unique dietary preferences, including gluten free, high protein, high fiber, lactose free, low calorie, low carb, low sodium, low sugar, low fat, organic, pescatarian, preservative free, vegan and vegetarian, Instacart said. Through Smart Shop preferences, customers can explicitly share household details, like whether they’re also shopping for a baby, toddler or pet. All of these inputs enhance Smart Shop personalization, helping surface items and aisles on Instacart that better align with each person’s dietary and household preferences.
New Health Tags — According to a recent Instacart customer survey, more than 70% of users have at least one dietary preference. To help them navigate grocery options with ease, Instacart developed an AI-driven Health Tag system, which scans product data at scale to extract key nutrition characteristics. This multimodal AI platform leverages vision-language models to extract rich product attributes from text and images to enable rapid, high accuracy tagging of nutrition claims, ingredients and dietary labels. Using this model, Instacart said it has tagged over 1.3 billion data points across food and beverage products in its catalog and today, Health Tags provide transparent easy-to-digest details on approximately half a million items across the more than 100,000 stores on the Instacart platform. Instacart’s 30 Health Tags now include gluten free, grass fed, heart healthy, high fiber, high protein, keto, lactose free, minimally processed, pasture raised, preservative free, no artificial colors, no artificial flavors, non-GMO, vegetarian, vegan, wild caught and more, helping consumers easily view and filter items based on their preferences, discover new products and make more informed shopping decisions, the release said.
Inspiration Pages — Instacart is also introducing Inspiration Pages, curated destinations within the Instacart experience that make expert nutrition advice and recommended recipes shoppable. The first page launched in collaboration with the American Diabetes Association simplifies shopping for people living with prediabetes, diabetes and obesity — while also supporting anyone looking to make healthier grocery choices, the release said. By making trusted nutrition guidance easily accessible, Inspiration Pages help customers find foods that align with their health goals. Instacart said it plans to continue expanding Inspiration Pages, launching new shoppable experiences based on Smart Shop preferences like high fiber and high protein, and partnering with additional organizations to bring more expert-backed guidance onto the platform. These pages will provide consumers with curated product selections and tailored recommendations, making it even easier to find foods that align with their preferences and goals, the release said.

Through AI-powered personalization, advanced preference detection and rapid product attribute extraction, customers can use Instacart to seamlessly shop for products that align with their health goals, turning grocery shopping into a smarter, more intuitive experience, the release said.

“Instacart is making it easier for people to shop for nutritious foods that support their health goals and lifestyle. With Smart Shop technology and Health Tags, we’re giving consumers the power to personalize their experience, with tools to filter and discover the best options for their unique preferences,” said Sarah Mastrorocco, vice president and general manager of Instacart Health. “Our collaboration with the American Diabetes Association takes this a step further by turning expert nutrition advice into easy, shoppable solutions through our Inspiration Pages. Whether you’re managing a chronic condition like diabetes or simply looking to make more informed food choices, we’re here to help make grocery shopping simpler and more personalized.”

Instacart and the American Diabetes Association have also worked together to build a turnkey nutrition program for health care partners, using Instacart Health Fresh Funds , customized by the the association to cover foods informed by its Standards of Care in Diabetes , the release said. This new resource enables organizations like health plans and providers to adopt the Fresh Funds program and refer patients to the American Diabetes Association Inspiration Page for reliable, actionable nutrition guidance, making it simpler to integrate nutritious food into chronic disease treatment and management at scale.

Using both the association’s Inspiration Page and Fresh Funds, health care organizations can now seamlessly connect members and patients with the shoppable, trusted content that encourages practical, everyday food choices, Instacart said.

AI-powered FarmWise prepares for next chapter in ag robotics

Mon, 17 Mar 2025 18:43:59 GMT

FarmWise, an artificial intelligence and computer vision weeding technology company in the robotic precision weeding technology space is restructuring its business, CEO Tjarko Leifer told The Packer on March 14. The Salinas, Calif.-based company, which employs a staff of 50, is best known for its Vulcan next-generation, intra-row weeder and precision cultivator that launched in 2023.

“The need for solutions like Vulcan has grown as growers face mounting pressure from labor shortages, rising input costs and operational complexities, said Leifer. “The technology has proven its ability to deliver meaningful efficiency gains and cost savings in the field. But even with growing demand and a product that works, achieving sustainable, profitable growth remains a significant challenge in today’s market.”

Vulcan is really a three-in-one implement that does precision weeding, cultivation and thinning, says Leifer.

On the specialty crop side, the company’s Vulcan technology is employed primarily by growers of lettuce and brassicas in the coastal California and Arizona vegetable crop markets. More recently, FarmWise entered the processing tomato market in California’s Central Valley, a new segment that Leifer says is nearly as big as the leafy green and brassicas market on an acreage basis.

Leifer points to one customer — a 10,000-acre vegetable grower with farms in Yuma, Ariz. and Salinas — who saw significant savings and a reduction in labor costs.

“Implementing the FarmWise Vulcan has transformed how we approach weeding on our operation,” the vegetable grower said. “Over just one season, we reduced weeding costs by nearly $550,000 and eliminated the need for cultivator passes on 64% of the acres covered. The AI-powered technology has allowed us to scale operations more efficiently, reduce labor costs, and improve precision in ways that traditional methods couldn’t match. This technology has set the foundation for even greater growth and profitability in the future.”

The grower also reported its per-acre weeding costs dropped 38% in romaine and 15% in broccoli. Additionally, 64% of its acreage no longer needed cultivator passes, cutting fuel use and field wear. And machines achieved up to 100 acres weeded per week, with 1-3 acres per hour efficiency and no compromise on weed removal quality.
With such notable results, why was FarmWise unable to scale?

“The short version is that we haven’t — with the resources that we’ve been able to raise — been able to reach profitability,” said Leifer. “And that’s such a critical threshold to get to for any business.

FarmWise CEO Tjarko Leifer is hopeful about the next chapter for the agtech company.

(Photo courtesy of FarmWise)

“We’ve had some great milestones we’ve achieved in terms of a product that really works for farmers, that drives a return on investment, that has a payback period of less than two years, and that’s deployed commercially and reliable day-in and day-out,” he continued.

But the company has also faced challenges.

FarmWise says the time horizons have been challenging for the company, because while it sees demand for the technology, it needs a longer runway to prove itself and drive adoption.

Leifer says broader macroeconomic headwinds have also impacted ag equipment investment overall — from rising interest rates and policy uncertainty to challenges around labor and immigration.

“Agriculture is a business that rightfully values caution and real-world proof,” Leifer said. “Growers want to see new technology work on the ground before fully adopting it — and we respect that approach.” Next-gen tech isn’t alone in feeling these headwinds.

The CEO also said these headwinds aren’t unique to startups. Large equipment manufacturers have seen sales drop 30% to 50%, making it a particularly tough moment for capital-intensive innovations, he said.

“If you’re a 120-year-old company, it’s a cyclical downturn,” he said. “If you’re a startup, it’s a much steeper hill to climb with limited resources.”

Agtech is in need of fellow visionaries to succeed, and Leifer sees this as a moment for the industry to rally around innovation, while recognizing the path from early adoption to scale is never linear.

“Building a great product is only half the challenge,” he said. “The next chapter is building the distribution and support infrastructure to bring it to scale. That’s where partnerships become critical.”

FarmWise’s core product is the AI-powered Vulcan, which weeds, cultivates and thins vegetable crops like lettuces and brassicas.

(Photo courtesy of FarmWise)

Labor considerations

While agriculture, like many industries, is in a labor crisis, in some cases it’s been slow to adopt technology that would alleviate that challenge.

“Labor cost inflation, especially coming out of COVID, and also changes in laws around how ag overtime is treated in California, are pressing issues, but with the new administration’s approach to labor and immigration, people really don’t know if labor will become plentiful or if it’s going to be much harder to hire people,” said Leifer. “So, there’s a lot of things changing right now, and a bit of a wait-and-see attitude.

“It’s an existential question for many segments of farming, in particular vegetable farming segments where there’s a lot of labor costs per acre,” Leifer continued. “Those costs are going up. It’s getting harder to find people, and farmers are very interested in technology and solutions that help them address that.”

Leifer says even if growers aren’t looking to reduce labor, they’re still looking for solutions that allow them to farm on a larger scale with the skilled labor they already have.

“How can you upskill the needs you have so you can do more with the good people you have? It’s how do we keep agriculture profitable and healthy in the United States and competitive? We need to find ways to be more productive,” said Leifer.

With AI and computer vision robotics, there are efficiencies that can be delivered today that couldn’t before, he said.

Robotics addresses inputs

In addition to the labor side of the story, there’s also the chemical side.

“There’s a lot of talk about some of the herbicides that are used in broccoli being removed from the market,” said Leifer. “And as you broaden out to row crops, there’s the whole herbicide-resistance story in a lot of those cropping systems, which ultimately mean farmers are going to need new technologies. We’ve always thought that AI and robotics can play a really important role there.”

Robotic machinery allows growers to see and understand the crop in the field, differentiate between the crop and weeds, and then control in real time how the machine is behaving as it goes through the field. As such, robotics also offers the potential to reduce the use of chemistry, drive crop fertility and automate tasks that are still done manually, says Leifer.

“There’s natural herbicide resistance that’s gaining and farmers are going to need new solutions in a broad range of cropping systems,” he said. “The market is working on providing innovations for them and we believe one of those — a big one — has to do with robotics and agriculture.”

The road to adoption

FarmWise says its flagship three-bed, 80-inch unit had an MSRP of $645,000. The annual service and support package is an additional $45,000. Leifer says the savings driven by the unit pays for itself in under two years.

“It’s a very attractive proposition,” said Leifer. “If the machine works and it saves people money, people should be ripping this out of your hands and off the lot to put into their fields.”

But employing AI-driven tech is not as simple as flipping a switch or screwing in a light bulb.

“It takes someone who believes in it and sees their success in the organization tied to the success of making the program work,” he said. “It takes an ownership group that wants to empower somebody like that. So, there ends up being a lot of nuances to how an organization can successfully adopt technology.”

What’s next for FarmWise?

While the current FarmWise team is preparing to wind down operations by April 1, the company says it is actively pursuing strategic opportunities — including acquisition, partnership and technology transfer — to ensure the Vulcan platform continues to thrive.

“We’re proud to have proven that advanced robotics can deliver real value in one of the most complex environments — agriculture,” Leifer said. “This is just the start of what’s possible with this kind of technology.

“We’re in active discussions with potential partners and investors,” he added. “There’s a path forward here, and I’m hopeful the technology we’ve built will continue to scale and serve more growers in the future.”

FarmWise says it remains committed to supporting its current customers and ensuring continuity during the transition period — including access to service and spare parts.

“We know growers depend on this technology, and we’re doing everything we can to provide stability and a strong handoff,” Leifer said.

How AI is transforming greenhouse, field growing

Mon, 17 Mar 2025 12:44:26 GMT

Agriculture wasn’t originally part of Valeria Kogan’s life plan.

The founder and CEO of Fermata, a company using artificial intelligence to help farmers monitor plant health and prevent crop losses sustainably, Kogan’s background in mathematics and biotech led her to apply her expertise to agriculture after seeing the inefficiencies in manual plant scouting.

With a doctorate in bioinformatics, she began as a researcher and later led AI teams at biotech startups across Europe and the U.S. Before Fermata, she co-founded Smartomica, an AI-powered platform assisting oncologists in diagnosing cancer patients. Driven by her passion for using tech to solve critical problems, Kogan moved from cancer labs to farmlands to create Fermata’s Croptimus — a tool she believes can make crop care more efficient, sustainable and profitable for growers worldwide.

While working on solutions for human health, Kogan said someone working in a tomato greenhouse asked about AI and how it could relate to agriculture.

“I was curious, so I decided to go see them, and I was fascinated with commercial food production. When they shared with me the challenges they saw, it became clear that the task technically is very similar to what I had been doing in the medical space,” Kogan said. “So, for Fermata, I started with the idea of taking the technologies already out there [in the industry] and bringing them to help monitor the health of plants. That was five years ago — and here we are.”

Valeria Kogan is the founder and CEO of Fermata, a company using artificial intelligence to help farmers monitor plant health and prevent crop losses sustainably.

(Photo courtesy of Fermata)

Fermata’s proprietary platform, Croptimus, provides greenhouse and field growers with a continuous, 24/7 view of their crops through AI-enhanced monitoring using a simple off-the-shelf camera. By detecting pests and diseases earlier than human scouts, Croptimus reduces the need for costly and environmentally harmful pesticides while boosting yields sustainably, the company says. As agriculture faces a shortage of skilled labor, this round-the-clock AI solution offers consistent, tireless monitoring that learns and adapts over time to each unique crop and environment to deliver accurate, real-time insights, Kogan says.

Basically, Kogan says, the company is solving the problem of plant health.

“Right now, up to 30% of crops are lost due to pests and diseases,” she said. “This is a major source of food waste on the planet, and a substantial source of CO2 emissions as well.”

Manual scouting is the norm, Kogan says, where people walk through greenhouses or fields and look at every single leaf of every plant to identify if something goes wrong.

“Of course, being humans, we’re not built for these sorts of tasks, so Croptimus helps with this task,” she said.

Greenhouse growers purchase security cameras, install them in the greenhouses — about four per acre — Kogan says, then the cameras zoom into every plant and the AI processes the data.

Fermata’s proprietary platform, Croptimus, provides greenhouse and field growers with a continuous, 24/7 view of their crops through AI-enhanced monitoring using a simple off-the-shelf camera.

(Photo courtesy of Fermata)

“There are thousands of images that we collect, and it helps identify whether there are any pests, diseases, viruses, nutrient issues ... we have a list of more than 25 things that we are identifying like basic pests, spider mites, thrips diseases, mildew or other fungi or bacterial diseases,” she said. “And if there is something within these classes, then we present that to the grower, and that’s what the growers learns from our app.”

There are three major value proposition points, Kogan says. First is the reduction of labor that a grower needs for the scouting process.

“Our case studies show that up to 50% were able to reduce the amount of human involvement in scouting,” Kogan said.

“The second is the reduction of crop inputs — especially with biological protection being so expensive, and the price is growing,” she added. “It’s really important to know what happened and to localize the problem in time, and then you can save a lot on the crop protection.

“Last but not least is crop losses,” Kogan continued. “We make sure that they don’t get out of hand and there are no significant outbreaks.”

The technology itself is crop agnostic, but Kogan said the company needed to start with something, so it began with greenhouse vegetables.

Kogan says one success story is a grower named Yonathan, who was skeptical and wanted to test the technology.

“We started working with Yonathan and providing him data, she said. “In addition, he also had regular human scouts visit — an external company that comes to him once a week and goes through all the plants and provides him with the report. He was really happy with them, and he didn’t want us.”

Kogan says her company would put the reports together once a week — Croptimus versus the scouts’ reports — and compared them for two months.

“They were absolutely identical,” she said.

Yonathan was still skeptical and wondered why he was continuing with the AI monitoring, but one day Croptimus detected a fungal disease, Kogan says.

“Because he was skeptical, he was sure his team would find it as well, so he didn’t share the data with his team,” she said. “But they didn’t, and the next week they didn’t, or the week after. It took them three weeks to detect the fungal disease. This convinced him of the benefits, and he went from a skeptic to a believer.”

While the company is based in Tel Aviv, the technology is available worldwide, and Kogan recently spent time in Canada meeting with the Greenhouse Growers Association.

Looking toward the future, Kogan thinks two things will happen in the next five to 10 years.

“First is that we’ll add other items on top of pest and disease detection,” she said. “We’re already looking at pests, which means distinguishing them from other insects in the greenhouse. Since we’re tracking insects already, we’ll start tracking the population of pollinators or other beneficial insects. We see several other tasks that can be automated through automation vision too.

“The other direction, with how AI is developing now, using multiple sources of data can help give a more precise answer,” she added. “So, sometimes you can say it’s either this disease or that disease, but if you know what the climate is, it’s like a definitive answer. That’s why we also want to add, later on, the other sources of data to make sure that all our predictions are more accurate because our AI is not just an image, it’s a look at the bigger issue.”

Google’s Parent Company Alphabet Disperses Its Ag Tech Subsidiary

Thu, 12 Sep 2024 15:35:00 GMT

Announced earlier today, Mineral, Alphabet’s ag company, will wind down its operations.

“Mineral will no longer be an Alphabet company, and our technology will live on inside of leading agribusinesses where they can have maximum impact,” said Mineral CEO Elliott Grant in a blog post.

Mineral was founded in 2018 as part of X, the moonshot factory of Alphabet, and it had about 100 team members.

What did Mineral develop and build:

• an image database of more than 17 crops in every stage of growth in multiple environments.

• A four-wheeled semi-autonomous rover platform with multiple configurations and the core functionality as a data collection machine.

• in-field harvest analysis and post-harvest crop condition ratings for berry crops in partnership with Driscoll’s

• a geospatial analysis platform that has collected more than 450 million acres of farmland.

• Phenotyping databases and analysis

• And additional machine learning and artificial intelligence technologies.

Here’s a link to previous coverage about Mineral.

Driscoll’s has confirmed it will license the tech it partnered with Mineral to develop.

“Mineral had partnered closely with Driscoll’s - the world’s leading berry company - to develop AI tools to improve crop phenotyping, better forecast yields, optimize quality inspections, and reduce food waste in the supply chain. Some of the technologies we developed have now been transferred to Driscoll’s and will be integrated into their systems to help achieve their sustainability ambitions. Driscoll’s is the first agribusiness to receive Mineral technology, and is a first step towards ensuring that our breakthrough technologies achieve the greatest impact,” Grant said.

In closing he gave an analogy of the company’s name to the how it can be applied as a verb in the agricultural context:

“In soil science, mineralization is the process by which the nutrients in organic matter are released in a form that makes them available to the plants around them. I think this is a fitting metaphor for the new chapter of Mineral - as our technologies will be mobilized into the agriculture ecosystem, with the goal of making it more sustainable, and more resilient.”