Last year, a shipping container with the Reservoir Farms logo facing traffic showed up on highway 68 just outside of Salinas. Many looked up the newest AgTech addition to the valley and wondered what would be going on behind its fences. Last Wednesday, the Salinas public had their first official view of Reservoir Farms at its open house. Beagle Technologies, LUMO, TRIC Robotics, High Degree, and Bonsai all had booths educating attendees on the struggles farmers are facing and what they’re doing to help. Reservoir founder, Danny Bernstein, and Salinas mayor, Dennis Donohue, both spoke about the increasing necessity of AgTech and the great work the Reservoir members were doing.

Also in attendance was Salinas High’s robotics team, Steel Boot. In his speech, Bernstien talked about the importance of introducing the next generation to AgTech, offered them a $1,000 donation, and announced an extended donation to any other local high schools that were interested in coming to Reservoir. He talked about how farming is local and emphasized that companies like High Degree have already begun hiring people from the area.

Most of these startups already had some form of prototype before the creation of Reservoir. The value Reservoir offers is a complete fabrication shop made by Andros, as well as acres of California specialty crops designed to be used for trials. When these startups are in the prototype stage, the fields can be used to refine the technology. If a startup were working directly with a grower, any malfunctions or misinputs would affect a farmer’s margins, but at Reservoir, these shortcomings are expected. Once a startup is more fine-tuned, these same facilities can be used to gather data on return on investment, making Reservoir a one-stop-shop for all stages of a startup.

One of the groups that stood out was High Degree, a startup that uses steam to kill weeds and soilborne pathogens. It works at up to 14 inches of depth, requires no REIs and is completely safe to use in organic production. At the event, they were telling attendees that all three of their prototypes were made at Reservoir Farms. They essentially started from the ground up and used the facilities Reservoir provided to build three complex, functioning machines.  What was initially just an idea when the company started in 2025, is now being used on select farms and already helping growers.

This was the first of hopefully many opportunities for the public to step into Reservoir’s fences and put faces to the company. Many drive past the barn inscribed “Home of Agtech” every day on the way to or from work. Hopefully now they have a greater appreciation of what is happening inside.

Last week, we took a look at the overall $5B strategy for specialty crop AgTech automation. This week we are going to dive into some of the specific re-allocation requests and rationale.

First, let’s look at the Inflation Reduction Act (IRA) Climate-Smart Ag Funding, which represents $19.5B and is being used to increase existing conservation programs through NRCS through payments for practices that reduce GHG emissions, improve soil carbon, reduce nitrogen losses, or sequester carbon. The major spends are: (1) $8.5B on cost-share practice payments (EQIP); (2) $5B on regional partner-led conservation projects (non-profits, Universities – RCPP); (3) $3.3B on multi-year stewardship contracts for conservation; (4) $1.4B on wetlands easements (ACEP); (5) $1B on conservation technical assistance. While there is demand for the program (it’s a government subsidy – that should not surprise anyone), it provides mediocre climate accounting (at best) while providing no improvement to grower economics. The success metric is the carbon sequestered and the funds deployed. It is already being modified to allow reallocation to wider funding options.

Much of this money would be better spent  subsidizing grower automation purchases. Keep in mind that many of the climate smart ag programs are positioned as meant to help farmer’s economics. The reality is that for many of these programs, farmer success metrics aren’t measured at either baseline or incremental performance and are rarely used as success metrics when evaluating the programs. Automation spending success can be measured by evaluating the actual impact on labor and grower economics, as well as by job creation (economic development impact).

Second, let’s look at the USDA Partnerships for Climate-Smart Commodities, a 2022 Pilot Grant Program that is going to spend $3.1B on 141 projects for: (1) technical and financial aid to farmers; (2) monitoring, reporting, and verification; and (3) market development (premiums for “climate smart” products). The Trump Administration reframed the program as “farmer first” and added requirement that 65% of funds go to farmers (by putting in place lower admin cost caps). So how is this program doing? Well, the original USDA targets were 60,000 participating farms, 25M+ acres, 60M metric tons CO2e. The actual results (two years in are 14,000 farms, 3.2M acres, 400k metric tons). As with the IRA program, there are no farmer success metrics established as targets or measured. As above, the metrics for automation incentives are much easier to manage and tie directly to labor and job creation metrics. These are better targets for rural communities than emissions reductions.

Now let’s take a look at the model for building the AgTech incentives infrastructure to help support accelerating AgTech purchases. It already exists and is part of the Climate Smart Funding efforts of the last 10 years. It’s the electric (EV) tractor subsidies program that uses CARB (California Air Resources Board) payments. The first program is FARMER, which pays for EV tractors that replace diesel FARMER, requires proof that the replacement tractor was turned into scrap (particularly wasteful by the way – there are a lot of farming communities that could use old diesel tractors that are not going to be used in the US market anymore), and provide an average check size of $80,000 – $100,000. FARMER represents 85% of program payments through CARB. The second program is CORE, which pays to incent EV tractor purchases, represents 15% of program payments, and has an average check size of $50,000 – $60,000.

As an example, one of the program participants (companies that want to participate in CARB programs need to be vetted and selected), Monarch Tractor made EV tractors that received significant CARB dollars. The business model for Monarch was selling 40 horsepower (HP) tractors for $85,000 and receiving a $50,000 – $60,000 payment from CARB. Even after receiving $155M in VC funding and significant CARB funding to help support their EV tractors being purchased by growers, Monarch was unable to raise more capital and was acquired by Caterpillar after attempting to pivot from a tractor company to an IP company.

Now let’s look at the success metric for these two programs. The program win was based on emissions reductions based on the number of hours the EV tractor was driven. But here’s where the EV tractors meet the road. It’s very hard to prove these emissions because it appears that the system was largely based on an honor system of self-reporting and tops down economic models that are not verified via any bottoms up analysis. I have seen some of these EV tractors in their native habitat at growing operations. They are very rarely in use because of the lack of a compelling use case, but the same unused tractors are getting credit for success metrics because of the factors mentioned above.

Next, let’s look at the Regenerative Pilot Program, USDA’s $700 million program that was announced December 10, 2025. It is being run by NRCS as an FY2026 pilot that redirects/set-asides funding inside two existing conservation programs: $400 million through EQIP and $300 million through CSP. The stated goal is to move USDA conservation funding from isolated “practice-by-practice” payments toward whole-farm regenerative conservation plans focused on soil health, water management, and “natural vitality.” It is a large conservation delivery pilot layered on top of EQIP/CSP, with a stronger emphasis on whole-farm planning, soil-health testing, bundled practices, and public-private supply-chain partnerships.

To date, many of the success metrics for similar practice payment programs want credit for approving a budget and spending the money. You can excuse farmers for not being overly impressed with passing a budget and actually spending the money. That’s basically table scrapes for any grower operations team. The real key is measuring the success of the spend. For many pilots and launched programs around regenerative practices, the success is only measured to the allocation and spend stage. It never makes it to actual grower metrics that improve their economics. As above, reallocating some of this capital toward automation incentives provides a much better set of metrics for rural economies.

I am in the middle of the analysis around federal economic development grants. For now, know that there looks to be $4.7B – $5.2B a year in federal grants in programs like Community Project Funding (EDI) ($3.3B in FY 2024), Economic Development Admin ($1.1B), Revitalization Grants ($320M), and USDA + DOE Econ Dev ($800M). The next action item is to identify which of these apply to rural economies and how the success metrics are measured to see how competitive the economic development impact of automation incentives compares to existing programs.

So that is the summary of programs we are looking to re-allocate dollars from to support the growth of automation revenue to $5B annually.

The Western Growers Innovation Team is proud to announce that SAMI Robotics has become the first company selected to receive sponsorship through the new Western Growers Association (WGA) and Reservoir Farms partnership. This milestone represents more than support for a single AgTech company; it marks the launch of a new model designed to accelerate the commercialization of technologies that address some of specialty crop agriculture’s most pressing challenges.

The partnership between Western Growers and Reservoir Farms was established with a shared vision: creating an environment where innovative agricultural technologies can be evaluated under real commercial farming conditions while generating objective data that helps growers make informed decisions and provides technology developers with meaningful feedback. Rather than relying solely on demonstrations or pilot projects, the program is designed to produce measurable performance data that reflects the realities of commercial specialty crop production.

One of the first major initiatives under this partnership is a series of field demonstration events that bring growers, technology developers, researchers, and industry stakeholders together to evaluate technologies in side-by-side comparisons. These events go beyond traditional equipment showcases by incorporating replicated field layouts, pre-treatment evaluations, and standardized trial protocols that allow participants to observe technologies performing under identical conditions.

SAMI Robotics is the first company to participate through this sponsorship, showcasing its automated harvesting platform in front of growers and industry leaders. As labor availability continues to be one of the greatest challenges facing specialty crop agriculture, harvest automation remains one of the industry’s highest priorities. Programs like this give emerging companies an opportunity to validate their technology where it matters most—in commercial fields alongside the growers who will ultimately determine its value.

At Western Growers, we believe successful commercialization requires far more than innovative engineering. Technologies must demonstrate reliability, operational efficiency, ease of integration into existing farming practices, return on investment, and measurable value for growers. By creating opportunities for companies to collect objective field data while working directly with commercial farming operations, the Innovation Team helps bridge the gap between promising concepts and practical, scalable solutions.

The collaboration with Reservoir Farms represents an important evolution in how the Innovation Team supports both growers and AgTech companies. Reservoir Farms provides a commercial production environment where new technologies can be tested under realistic operating conditions, while Western Growers contributes its extensive grower network and experience conducting independent technology validation. Together, the partnership creates a platform that accelerates innovation while keeping grower needs at the center of the process.

The sponsorship awarded to SAMI Robotics recognizes companies with the potential to solve meaningful challenges facing specialty crop agriculture. As the program expands, additional sponsorship opportunities will be available for emerging companies developing technologies in harvesting, weeding, spraying, automation, artificial intelligence, sensing, and precision agriculture.

This initiative reflects Western Growers’ ongoing commitment to helping its members identify technologies that can improve labor efficiency, increase productivity, strengthen sustainability, and enhance long-term competitiveness. Equally important, it provides growers with transparent, objective information to support future technology investment decisions.

The Western Growers Innovation Team thanks Reservoir Farms for its partnership and congratulates SAMI Robotics on becoming the first recipient of the WGA–Reservoir Farms sponsorship. This is just the beginning of a collaborative effort to bring growers, innovators, and researchers together to validate the next generation of agricultural technologies and accelerate their adoption across specialty crop agriculture.

Because Salinas Valley can benefit from the progress Silicon Valley is making on AI and this event will help both sides recognize that.

There have been a lot of (and many are ongoing) discussions about the interplay between Silicon Valley and Salinas Valley. Lately, Silicon Valley is heavily focused on AI – how it’s getting used, how many vibe coders it is enabling, how many jobs it’s going to create, which industries may lose a lot of jobs, and the amazing hockey stick revenue growth of two of the largest “startups” (hard to use that word without quotes when revenue gets to the billions … hello OpenAI and Anthropic) that most of us have literally never seen before. That part is awesome. Better AI tools enable all tech startups to build faster, rollout better, and scale more aggressively while helping growers get access to AgTech faster because it is using AI. In a Silicon Valley world of “live fast, break glass, try everything – and report back by lunch” AI is like an adrenaline IV for tech developer teams.

Meanwhile, in Salinas Valley, it’s less about hockey sticks and more about boring old spreadsheets, because in order for someone to start buying robots, the math has to math. It’s bad to break glass or too many other things, and one failed experiment may mean you don’t get a second or third one without some serious back pedaling and a great explanation for why that happened … that the grower operations team believes (and that does not depend on you showing remorse for the failed test.)

While there’s plenty of dialogue on the funding fit side of things (some of my investor friends say Silicon Valley funders need to figure out how to tweak their model to AgTech. I’m of the opposite view that Salinas Valley needs to figure out how to grow companies that can scale on a growth curve that looks more like other tech segments because Silicon Valley is doing just fine creating unicorns for the most part and doesn’t need to tweak the model for an under-performing segment relative to its peer group), there is complete agreement on the product side of things.

AI is changing everything in all segments and Silicon Valley is leading the way. OpenAI, Anthropic, Google Gemini, Facebook Llama, and Apple’s (well, whatever Apple ends up doing with AI – some day!) are all in the heart of Silicon Valley, as are many of the AI investors. All of the advancements in AI can and will eventually benefit vertical markets like AgTech. It has always been this way. The platforms and infrastructure build and scale and then the verticals follow, leveraging what the horizontal players learned and built.

Enter Physical AI – the cross-over point between the two valleys. Yes, Nvidia (and now AMD and Intel and a bunch of others) is building chips and hardware for AI performance across several key functions (inference, training, etc.) to increase and to drive token capabilities up while driving cost per token down so the subscription plans can bump up in price as margins hopefully go up because token prices plummet. In theory, a win for everybody. In practice, well we will see, and mileage may vary. And every other Tuesday a new model arrives from one of the big dogs that create awesome new capabilities that all users get to take advantage of right now. So, the pace of change is accelerating and tools and infrastructure are getting better at the same time.

Now we go over to AgTech where hardware means tractors and robots and hardware and software have to work 24/7/365 in obscenely tough conditions – outdoors, all kinds of weather, not many days off, and it better not break down very often and get fixed quick when it does. Other than that…well, you get the idea. The part where Salinas Valley needs to pay attention is to watch the growth of the platform components and figure out how you can do Physical AI better and faster by leveraging the Silicon Valley toolkit and using it to speed up your AgTech automation solution.

Enter Ruggedize – the first show designed to bring those exact two audiences to the table in the same place at the same time. Naturally, it happens in Salinas at Reservoir Farms next month – this is the first one, and I predict it will be a big success because both sides will benefit from the panels, conversations, and lessons exchanged and learned over two days. When people ask me what’s the benefit of having someone like Danny Bernstein come down from Silicon Valley to found Reservoir Farms, Ruggedize is a great example of exactly the benefit – someone who knows both Valleys well is building an event that both Valleys will get a win from.

Ruggedize information here.

The most important number in specialty crop AgTech is annual automation revenue, and the big target we are taking on is how we can get that number up to $5B in one year. I’ve written before on the importance of getting automation to become a category with over $1B in annual sales. That metric is a big one from a segment perspective because in many tech categories there is precedent for that level of revenue generating follow-on segments of significance. In the case of automation, the opportunities are: (1) systems integration (helping ag operators incorporate the robots into their farming operations); (2) data/analytics (taking the data from multiple robots and turning it into a recommendation engine for agronomic decision making on future rotations; and (3) manufacturing/service ecosystem (at a certain level it becomes it worth it for manufacturers to begin manufacturing equipment close to their customers, no matter where HQ is located). In the case of enterprise data centers, as various iterations of storage (NAS, SAN – basically any combination of storage (S) and networking (N) you can acronym) reached $1B, the supporting infrastructure around systems integration and data/analytics tools often got to be 25-50% as large as the underlying $1B category.

So then I started doing some analysis on what the projections for automation revenue could look like over the next 5-10 years and beyond. As I dug into the amounts of venture capital already invested, the amount and trajectory of current revenue from known players in the space, and the likely trends of VC and revenue going forward, two things became apparent. First, with just the known players and expected growth trajectory, AgTech automation is poised to go from ~$310-325M in 2025 to $1B in 2030. This is a 26% compound annual growth rate (CAGR), and the growth rate for the last couple of years has been at least 25-30% with some spike years. Second, if you look at the expected AgriFoodTech VC funding for the next 5 years and the expected percentage on automation, it is very likely that a clear path to $2.5 billion in the next 10 years is not just possible but approaching probable.

How does the $1B show up? Well, over the last 10 years there has been $2.8-$3.2B in automation investment, and that has generated the momentum to reach $1B in revenue by 2030. This will largely come from non-harvest automation, including weeding robots from Carbon Robotics (laser weeders) and Stout (mechanical weeders); spraying robots from GUSS, Ecorobotix, and Verdant; thinning robots from Niqo (and worth noting that some of the weeding robots can also thin – and vice versa over time), harvest assist (Burro), and autonomous mobility platforms (Bonsai/Farm-NG and Agtonomy). These players can get most of the growth to $1B without new entrants – any new entrants (and we do expect some) will accelerate the growth rate and bring the target data for $1B into play earlier. So that is the clear glide path to $1B category status in 5 years.

Now let’s look at the path to $2.5B. Even with AgriFoodTech venture capital dropping 70% in 4 years ($54B in 2021 to $16B in 2025), we are still at $16B for 2025 after $16-17B the past two years. For the moment, it’s fair to model the next 5 years of VC at $15B per year (in range lower than the last 3 years in case there is further reduction). This means we can expect the total AgriFoodTech VC number for the next 5 years to land around $75B. Then, let’s look at the automation percentage of that number to see where we should forecast the investment for the category. It should be no surprise that once the funding for vertical farming and alt-proteins largely stopped (they turned 42% of the $54B in 2021 into a dumpster fire and stopped getting checks – or at least got a lot fewer – as a result), the percentage that invested in categories solving real problems started to increase. Automation, which is solving the real problem of labor, went from 1.6% to 5.6%. If I had to bet, I would bet that the percentage invested in automation is more likely to go up than down in the next 5 years. For modeling purposes, I used a flat line allocation of 5.6% of the $75B. That gets to $4.2B over 5 years that is forecast to be spent on automation.

Now we can make some assumptions about the impact of that $4.2B based on what we have seen from other segments in the past. We know that capital efficiency increases for a segment when it moves from going from $0 to $1B to $1B to $2.5B. This makes sense. The first billion in revenue supports the build-out of infrastructure and the next billion (and beyond) are able to leverage the installed base and ecosystem built for getting the first billion created. Recall that the clear path to the first $1B in revenue was based on ~$3B in AgriFoodTech VC. Based on what we know about history, it’s easy to model the next $4.2B as more than capable of generating an incremental $1.4-$1.5B (if you just straight-line the $3B result to $1B, $4.2B would expect to generate $1.4B – I think there is upside because any new progress from automation startups from the first $3B will capture upside beyond the model. For these reasons, I believe that if the $4.2B in automation revenue provides a clear path to $2.5B in annual automation revenue.

Now that we have a clear path to $1B in revenue based on $3B in investment and an equally clear path emerging to $2.5B in revenue based on the next forecast of $4.2B in investment, the clear next question is what do we need to do to get to $5B in revenue. And a supporting question is what does that number enable in terms of new business opportunities, and, even more importantly, what does it mean in terms of job creation?

First, let’s look at the capital we need to get to $5B. Based on the fact that $3B in VC resulted in a clear path to $1B in annual revenue and $4.2B in VC should get us to $2.5B, it makes sense that less than $7.2B (the number modeled for the first $2.5B) in VC to get to the next $2.5B in revenue. But we have to be honest about the VC space, particularly in AgriFoodTech. While AI remains a high-flyer and solid performers in cybersecurity and fintech continue to emerge and get good outcomes, AgTech remains a long slog with limited exits, almost no IPOs, and very few M&A transactions. So the entire extent of VC for automation may be the $4.2B for the next 5 years we already have modeled in. For the rest of the exercise, I assumed we would need additional non-VC sources of capital to fund the additional $2.5B in activity.

So what are these new sources? I have been looking at both potential capital sources and programs for re-allocating spending from other programs to AgTech automation. I have found some programs that are using poor metrics and/or not measuring their success (or failure) very well. Three large areas of potential re-allocation that have been identified so far are climate-smart funding, regenerative ag practice payments, and federal economic development grants. I’ll dive into each in a future article. For now, I just want to summarize the high-level findings. I identified $23.8B in federal funding that are aimed at agricultural operations: (1) $19.5B in IRA (Inflation Reduction Act) climate-smart ag funding; (2) $3.1B in USDA Climate Smart Commodities; and (3) $700M USDA Regenerative Ag Pilot Program.

If we re-allocated 5% of that total to provide incentives for AgTech automation investment, we would have $1.1B in incentive payments. If you turn those incentives into an equipment repayment program similar to the EV tractor programs FARMER and CORE (which are run by CARB – the California Air Resources Board) and provide 40% public funding payments to support the 60% made the private funders (the growers buying the equipment), you get an additional $2.9B in additional automation revenue (which is comfortably over the $2.5B needed to reach $5B). We already have equipment support purchase programs for agriculture and other industries. We can re-use that infrastructure to accelerate faster towards the $5B target. We do not need VC funding – equipment funding can help reach the same goal.

And now we get to the point of the exercise – what does it mean when we get to $5B in automation revenue for specialty crop AgTech? I have been digging into this from a few viewpoints. The short version is we get three great outcomes at the same time:

1. There are 67,000 new jobs created that significantly over-index in rural communities: (1) 42,000 – 50,000 direct jobs at the startups that are designing, building, selling, and supporting the robots; and (2) 25,000-40,000 indirect jobs in related industries (like AgTech dealerships or manufacturing firms). This is why we believe that at $5B in automation revenue, this effort becomes rural economic infrastructure all across the US in places where specialty crops are grown.
2. Between 20-25% of the total US farm labor hours are automated. Much of this is continued growth in the non-harvest segments discussed above. We are also modeling some (but limited) progress in harvest automation. This will help reduce the pressure on farmer operations from increased H-2A immigrant farm workers. This is a significant percentage of the ongoing labor challenges that get solved through the use of automation solutions.
3. Based on what we have seen from tech segments historically, when a primary category like automation reaches $1B, there are sub-categories that tend to emerge. When we can push that category number to $5B, the sub-categories increase in opportunity size and in this case increase in count. In the case of automation, I believe there are 5 sub-categories that get created at $5B in annual revenue: (1) systems/operations integration – the dealers and support network that help the robots work inside of and integrate with grower operations represent a $1-2B annual opportunity; (2) the data/analytics tools that emerge from all the data captured by the robots that are making regular passes through fields also represent a $1-2B annual opportunity (this is software solutions – no hardware required); (3) manufacturing growth inside of specialty crop growing regions (we are already seen home grown manufacturing in US specialty crop areas as well as additional manufacturing from international startups seeing success and traction in the US market; (4) bio-circular economy solutions (the re-use and recycling of bio-mass from permanent crop acreage – trees and vines – that are being pulled out for market and SGMA water risk considerations and from the food production facility waste streams in many grower operations that produce things like bagged salads; and (5) AI workflow optimization – which will certainly increase operational efficiencies across the agriculture supply chain based on what we are already seeing from organizations in other segments the past 3+ years. Together, these 5 represent $3.5-$5.5B in annual opportunity in addition to the underlying $5B in automation.

There’s the summary of the strategy around automation – existing VC investment pushes us to $1B in annual revenue and expected VC investment pushes us to $2.5B, then $1B in re-allocated capital supports an incremental $2.5B based on $1B in investment for an equipment support program with 40% subsidies on selected automation equipment. That $5B then creates the 3 outcomes above: (1) 67,000-90,000 jobs; (2) 20-25% of labor automated; and (3) the 5 sub-segments above, which represent $3.5-$5.5B in known adjacent annual opportunity.  We will be working hard with our partners in agriculture, AgTech, DC, and Sacramento to dial in a lot of the details the next several years.

There is more to be done to continue building this out. We need to continue tracking progress towards the $1B in 2030 and $2.5B in 2035, as well as the VC funding for automation to support it. We need more and better solutions around harvest automation to round out the portfolio and drive revenue targets even higher. We need to model the path to $5B to make sure that the startups on the field and emerging soon can get to the kind of manufacturing volume we need to support the $5B target. Lastly, we need to do a fairly complete re-think on the strategy for Universities. To me, this is one of the largest opportunities. Our ability to change the strategy from primarily research deliverables to commercialization deliverables is job 1, and to do that we need to re-allocate dollars toward increased investment in IP protection, licensing, and start creation and we need to incent those outcomes at equal importance to research. We need a healthy front end of the funnel for innovation to create the number of startups required to get enough successfully commercializing to get to the $5B number. More on these topics later. For now, we are excited to share the outline for what we are working on in the WG Innovation team and what we’re building for over the next 10 years.

Water is once again right in the middle of a hot California dispute!

Kudos to Erik Benson for a great article on the market dynamics that are impacting both agriculture and AI data centers around water rights, land values, and the relative impacts on state economies that are or are considering the impact if both are increased or decreased. This is not a hypothetical exercise. Data centers are indeed looking for land, and some of it is or has been used for agriculture. Erik does a great job of diving into the details. I recommend giving the full article a read. It’s worth it because of the time he took to frame things up and reality check a few assumptions many of us that cross both ag and AI ecosystems end up buying into without always wrapping the whole story and context around it.

I think this topic deserves a full response (but that will take me a while!) For now, here are my quick thoughts to provide a little more context around the water conversation:

1) The SGMA allocations around groundwater are getting real this year as allocations from Groundwater Sustainability Agencies start passing out groundwater allocations that will require farmers to fallow acreage in high-risk groundwater basins. In some cases, depending on the type of water rights and basin classification risks, farmers are already making decisions to move from permanent crops (tree crops like nuts and stone fruits or vine crops like wine grapes and table grapes) to rotational crops to mitigate the risk of a fallow allocation from SGMA.

California could have avoided the looming crisis that is SGMA by building more surface water storage and conveyance (Sites Reservoir and many others like it, supported by canals and aqua-ducts) or desalination plants. The solution for a supply shortage is often building more storage. See Peru­—$24B in government commitments for 22 water storage projects— as the opposite approach to California.

2) The water wars were already happening before AI data centers. The water costs and supply challenges were already pushing agricultural acreage toward the highest and best use. Combine rising water costs and supply challenges with the overall regulatory landscape of California agriculture and the never-ending labor cost increases and farmers needed to continually move to better and better economics well before AI became a buzzword.

Crops will now compete with AI data centers just like they competed against other agricultural uses the past couple of decades. In that sense, everything old is new again, and where AI provides the highest and best use it will take some of the agricultural land and convert it.

Link to Erik’s article – (19) Dirty Water | LinkedIn

The Next Gen. Ag. Worker Program grant started back up this April.  This round of funding is through the California Governor’s Office of Business and Economic Development (GO-Biz) and will run for two years.  It will continue to support the Internship Reimbursement Program, Ag Tech X Ed events at community colleges, the LinkedAg.com website, and workforce development initiatives that connect growers with local community colleges and four-year universities.

Internships are in full swing with nearly 45 students working in ag tech internships throughout the state.  Interns are community colleges and four-year university students, and employers are providing opportunities for students to have hands-on learning experiences in the field, office, laboratory, and plant. There is a $3,000 reimbursement stipend paid to the employer for each ag tech intern once the internship is completed.  There are still a few spots left.

If you currently have interns and would like to inquire about the program, please contact Carrie Peterson at [email protected].  (Please note that there is no final “N” in her last name in the email address.)

On June 12, 2026, Danny Bernstein of Reservoir Farms made an X post featuring a field demonstration by TRIC Robotics. Chaos ensued. Within a few hours of the post going up, it had millions of views, thousands of reposts, and hundreds of comments. In these comments, there was plenty of support, but users questioned safety, effectiveness, and trust in AI and machines in farming.

“@POTUS, We demand these on every farm. NO MORE CHEMICALS. MAKE IT HAPPEN,” @scdlcaramia.

“This is what decimating the night time pollinator population looks like,” apparent entomologist @FedUpInTheMid.

“AI will replace jobs in every industry on the globe,” @MartiniGuyYT.

“Plants need dark to respire and facilitate flowering hormone triggers, here you are with a UV, Genius,” @Fugaziplacebo.

“Don’t they get UV light during the day (from the sun)??!” @Justcuriousjta.

These are just some of the hundreds of comments that reveal the disconnect between the work that startups like TRIC are doing and general population awareness. Current trends and initiatives like MAHA are leading people to seek healthy living practices and pesticide-free produce, but consumers remain distrustful of technology. Now more than ever, there is a desire to be connected to the processes of growing food.

When something from the agtech world is successful enough to be seen by the general audience and becomes part of the zeitgeist of current ag trends, it offers some useful insight into the consumer. Capturing their attention also provides an opportunity to educate and help bridge the gaps between those growing food and those buying it. Most of the comments listed above are made in good faith, but many of them are thrown into the void of the internet and never taken advantage of and followed up on. At best, people ask grok, which is a form of education, but lacks the nuances that a company like TRIC could share about their own technology.

What do we do with this information?

Despite the sceptics, this post’s virality was a good thing. Danny Bernstein, TRIC Robotics, and Reservoir Farms follower counts skyrocketed, creating an audience for future posts. 6 million people have seen the post, and if you look up TRIC Robotics on X, there are at least a dozen copycat posts, working to do free advertising for TRIC.

The phrase “all publicity is good publicity” is certainly not true. Every year we see companies’ reputations torn apart by poor choices from their executives or media people.

Figures in the ag tech ecosystem must be intentional with what they share and how they share it. Media needs to be something that is monitored, measured, followed up on, and most importantly, constant. Companies must maintain a flow of communication with both their clients and the general public to build a reputable brand. One successful post with no follow up actions is just a blip in the X timeline, but post after post after post has the potential to give a startup real staying power and attract the attention of those that they want to attract.

Over the past several years, specialty crop growers have seen an increasing number of new spraying technologies enter the market. From electrostatic systems to retrofit spray enhancement technologies, manufacturers continue to make claims around improved coverage, reduced inputs, and increased efficiency. While many of these innovations show promise, growers consistently ask one simple question before making an investment:

Does it kill bugs?

That question became the driving force behind a new research initiative led by the Western Growers Innovation and Science Teams.

Last week, Western Growers Innovation partnered with researchers from UC Davis, led by Dr. Ian Grettenberger, to launch the first of several independent field trials focused on evaluating insect control performance of emerging spray technologies under commercial production conditions.

Rather than relying on manufacturer data or laboratory studies, the objective is straightforward: generate independent, university-led data that helps growers make more informed equipment decisions.

Working alongside Mark Mason of Huntington Farms, who helped identify the most important validation criteria from a grower’s perspective, the research team designed a repeatable field protocol centered around the one metric growers care about most—effective insect control.

The trial compared three different spray approaches operating under commercial conditions:

• A conventional broadcast boom sprayer
• An on-target electrostatic spraying system
• A MagrowTec retrofit spray enhancement system

Before any applications were made, the UC Davis team established replicated treatment plots, verified diamondback moth pressure across the field, and mapped treatment zones to ensure the study would produce statistically meaningful results. Each participating technology then completed its application according to the research protocol, allowing researchers to evaluate performance using consistent scientific methods rather than anecdotal observations.

With applications now complete, researchers are monitoring the trial and collecting post-treatment data to quantify differences in insect control across the three systems.

While these results will provide valuable insight, Western Growers views this study as only the beginning.

Pest pressure, weather, crop growth stage, and environmental conditions can all influence spray performance throughout the season. For that reason, the Western Growers Innovation Team will continue conducting additional trials over multiple application windows to build a more comprehensive dataset before drawing conclusions.

This approach reflects a broader commitment by Western Growers to generate practical, grower-focused information that reduces uncertainty around technology adoption. Independent validation from trusted researchers gives growers greater confidence when evaluating new technologies.

A big thanks to On Target Spray Systems and MagrowTec for their willingness to participate in this independent research effort. Advancing agricultural innovation requires collaboration between growers, researchers, and technology developers, and their participation helps generate the unbiased data that benefits the entire specialty crop industry.

Once multiple trials have been completed and the data has been analyzed, Western Growers plans to publish the findings as a publicly available resource for the specialty crop industry. By making these results broadly accessible, growers will have objective information they can use to compare technologies, evaluate return on investment, and make decisions that best fit their operations.

Innovation in agriculture isn’t simply about introducing new technology—it’s about generating trusted data that helps growers adopt the right technology with confidence. Through continued collaboration with leading researchers, progressive growers, and technology developers, Western Growers is helping build the independent evidence needed to accelerate practical innovation across specialty crop production.

It was a great week at the Salinas Biological Summit last week at the Salinas Rodeo Grounds. A full day of workshops and a VIP gathering on Monday, a full day of bio-controls content on Tuesday, and a full day of bio-stimulants on Wednesday. From a content perspective, I believe it was the best Summit yet. One of the topics that continually came up in content and conversations was the need for more tests for bio-controls to measure their results relative to pesticides, herbicides, and fungicides.

I was talking to Pam Marrone about this topic. As one of the pre-eminent founders and advisors in the biologicals space, few are more familiar with the requirements and costs and processes for successful tests than Pam. I asked her what she thought the current cost for a test was for a bio-control and she estimated $25,000-$30,000. That is in line with what we have seen at Western Growers from a variety of Contract Research Organizations (CROs). I then asked how many tests the average product required to get to registration and ready for commercialization. She estimated 80-100 and leaned toward 100 as the right number because products need to be tested in a wide variety of soil, water, and crop conditions.

So basic math suggests that $30,000 per test times 100 tests is … $3,000,000. That means that every product (and startups are usually built to go beyond one product after the first one works and begins to commercialize), a startup will need to come up with $2,500,000 – $3,000,000 just for testing before they can raise an A round. Let’s walk through the expectations for general startups and then for AgTech startups.

For most startups (any tech segment), the fundraising progression and expectations looks like this:

Pre-seed: Can this be real?
Series A: Is it working?
Series B: Can it scale?

For AgTech, it’s a slightly different:

Pre-seed: Can this solve a specific farm problem?
Series A: Does it work in real commercial field conditions and produce measurable grower ROI?
Series B: Can it be deployed, supported, and expanded at scale with improving unit economics?

Among the challenges an AgTech startup must address before raising an A round is whether you can solve a specific farm problem. In this case, that means proving that the bio-control can solve the specific problem of replacing a chemical application that has been restricted or banned. To prove that, the startup has to get through the trials process mentioned above. This means that before most startups get to an A round, they need to complete 100 trials and come up with $2.5-$3.0 million in capital.

I believe the AgTech ecosystem needs to work on developing platforms that can help reduce both the number of trials required for registration and the cost of each trial. Pam and I and others are going to start working on some strategic options for reducing both numbers. Here are my early thoughts on that process:

• Reducing the cost of each test

There are hard costs associated with each test, and the $25,000 – $30,000 number includes the cost of preparing the test acreage, planting it, growing it, and harvesting it so that the ground is put back into it’s original state before the trial. Many of the tests are done by Contract Research Organizations (CROs) and can happen on a variety of acreage types, depending on what the requirements are for each test. Western Growers works with multiple CROs, and that cost range is what we are used to paying with some variance due to complexity or unexpected challenges of a particular test.

I believe there are two primary options for reducing the per-test price. First, you can secure testing across a large number of acres to support a large number of tests. This will reduce the per test cost. That of course requires access to acreage that can be leased or purchased and used partially to primarily for testing purposes. This will require significant capital. Second, you can reduce the price by finding a partner that can help subsidize the testing cost. For example, would a genetics company be interested in subsidizing the costs to reduce the costs to the bio-control startup or to many separate bio-control startups?

• Reducing the number of tests required

Reducing the number of required tests may prove harder. Needing to do 100 tests is not an absolute hard-line rule but it is a number that many subject matter experts in the field have mentioned (sometimes as a range – i.e. 80-100 tests need to be run). I need to dive into the rationale for the 100 number. My limited understanding so far is that you need to test both for the results you hope for (i.e. that the bio-control does indeed act as a pesticide alternative) in different circumstances (i.e. different soil conditions, water conditions, weather conditions) and than make sure you do not deliver any unintended consequences that can be tied back to the product being tested. So it’s a two-sided test – does it do what it’s supposed to do in multiple types of conditions and does it avoid doing things you don’t want it to do in similar conditions?

Some of the effort to reduce the number of tests required would involve re-examining the required test conditions and trying to maintain a reasonable confidence level while using fewer tests. The other option is to get a platform partner to support some of the trial costs under the theory that if more startups make it because they underwrite the tests, there will be more options to sell down the road, and that can be more effective than purely relying on internal R&D efforts.

The bottom line is this – we need to figure out a way to reduce the per product testing costs from $2.5 – $3.0 million to something 40-60% less than that. This will reduce the cost burden on startups on getting to their first product and getting through the certification process. I’ll be writing more about this objective as we make progress. In short, what we are doing with Reservoir Farms to reduce automation MVP time and capital requirements by 30-50% (or more) also needs to happen with biological (specifically bio-controls) startups.