Pilot Valley

One of the key takeaways from our Intelligent Ag event in São Paulo last week was that more than 90% of AI pilots in agribusiness never make it to adoption.

This doesn’t mean the technology is bad, corporates are not interested, or startups are building the wrong things. In many cases, the problem is more basic: the pilot wasn't set up with a clear enough path to become part of the business.

AI is already being used in agriculture for remote sensing, satellite imagery, weather forecasting, pest and disease detection, precision agriculture and machinery automation. These are no longer futuristic use cases. They are already being tested and deployed across the sector.

So the real question becomes: if AI is already useful, available and increasingly proven, why do so many pilots still fail to become part of the business?

Our keynote speaker Jose Damico of SciCrop called it the “valley between pilot and scale.” For most startups, the valley of death usually appears when they try to scale. In AgTech, the first valley often comes earlier. It sits between pilot and customer adoption.

The second panel, moderated by Mauricio Infantini from CNH, kept coming back to the same issue. Corporates, startups and investors are no longer talking only about technology. They are talking about data readiness, internal ownership, ROI, integration, governance and culture. Less sparkle, more plumbing. Less demo, more delivery.

Agriculture understands preparation better than most sectors. Nobody plants because the seed brochure looks good. You check the soil, the machinery, the forecast, the financing, the team and the window. AI pilots deserve the same discipline.

No Auto-Pilot

Corporate innovation has a weakness for pilots. They are visible, contained and easy to announce. A business unit has a problem. A startup has a solution. A project is launched. Everyone can point to movement.

But AI adoption does not run on auto-pilot.

Pilots in agriculture are unusually hard because the field does not wait for the project plan. Seasonality limits how often a solution can be tested. A crop cycle does not wait for procurement. A pest outbreak does not wait for legal. A logistics bottleneck does not wait for the next steering committee. Miss the window, and the company may lose months before it can try again.

Then come the practical issues. Field teams may not buy in. Data may be messy or inaccessible. Connectivity may fail. The real operation may not behave like the neat version described in the kickoff meeting. The startup may discover that the problem is different from the one it was asked to solve.

That is why a pilot can work technically and still fail commercially. It can produce a dashboard nobody opens, a prediction that arrives too late, or an insight that does not connect to a decision. It can excite the innovation team but never reach the operational team that owns the budget, the people or the process.

The usual assumption is that the main risk is technological. Will the model work? Will the AI produce a better answer? Those questions matter, but they are rarely enough. In agribusiness, the bigger risks are often organizational and operational: who owns the result, which workflow changes, what metric justifies expansion, and who has authority to move from test to rollout.

Without those answers, even strong technology can stall. The pilot may generate activity, but activity is cheap. Adoption requires time, trust, budget, behavior change and internal alignment.

Built In, Not Bolted On

AI only matters in agribusiness when it changes a decision.

Many companies are still testing tools around the edges of the business: LLMs, dashboards, agents, image recognition, predictive models and workflow automation. Some create useful quick wins, especially when they remove manual work or speed up analysis. But too many remain disconnected from the core systems and daily decisions that actually drive the business.

The symptoms are easy to recognize: isolated projects, uncertain ROI, shadow IT and no data integration. AI becomes something to test on the side, rather than something that changes how the company operates.

That is where many pilots lose momentum. They prove that something can be analyzed, predicted or automated, but they do not prove that the result will be used. A dashboard nobody opens does not change much. A recommendation that arrives after the decision has been made is just a very smart post-mortem.

Agriculture is full of decisions where better intelligence can make a real difference: when to spray, how to price credit, which freight route to choose, or which producer needs support before a small issue becomes a larger loss. AI can help with these decisions, but only if it sits close enough to the workflow to influence action.

The better question is no longer “What AI tool should we test?” It is “Which business decision would become faster, cheaper, safer or more valuable if better intelligence were built into it?”

ROI Roots

For an AI pilot to become part of the business, it needs to prove ROI. Everything else is a nice-to-have.

That may sound obvious, yet many pilots still begin with fuzzy goals like “improve efficiency,” “generate insights,” or “support ESG.” These can be useful directions, but they rarely close a budget discussion. In agribusiness, value needs to show up in familiar terms: reais per hectare, sacks per hectare, input reduction, days saved, risk avoided, freight costs lowered, claims reduced, working capital released or field visits optimized.

This is where Damico’s three pillars of data, integration and culture become practical. Data gives the model something useful to work with, but only if it is organized, accessible and tied to the real problem. Integration matters because an AI output sitting outside the workflow usually becomes another report. Culture matters because field teams, managers and operators need to trust the recommendation enough to act on it.

That trust is especially important in agriculture, where many decisions still depend on local knowledge and experience. A model may detect a pattern, but someone still needs to believe it enough to change a route, adjust an application, approve a credit line or call a farmer before a small problem becomes a larger loss.

ROI is also a useful discipline because it forces clarity. It makes the company define what the pilot is supposed to change, who needs to use it, which data matters, and whether the result is strong enough to justify moving forward.

The more concrete the metric, the easier it is to decide whether a pilot deserves to move forward.

Pilot Prep

A better AI pilot starts before anyone chooses a startup.

The first step is to define the problem properly. What decision needs to improve? Who owns that decision? What is the cost of getting it wrong today? Too many pilots begin with broad ambition, such as “use AI for ESG” or “improve productivity,” when they need a sharper business question.

The second step is to understand the solution landscape. Many corporates know they want AI, but have little visibility on what is already available, which startups are credible, and which tools are mature enough for their context. A good pre-pilot process should help separate real solutions from impressive demos.

The third step is to design the pilot for adoption from day one. That means defining success metrics, data requirements, internal owners, user workflows and next steps before the test begins. A pilot should not end with everyone asking, “Now what?”

We are developing a new initiative, the Pilot Readiness Program, to help agribusiness companies prepare AI pilots before they start, with a focus on problem definition, solution curation, internal alignment and ROI targets. More on that soon. For now, the point is simple: AI in agro does not need more pilots. It needs better-prepared ones.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Green Gap

One thing is universal about farmers: they are doers, not sayers. They do most of their talking in the field. They plant, test, adjust, harvest and deliver the best product they can. In general, they expect quality to do the talking, and they expect the market to pay accordingly.

That instinct makes sense in agriculture, where results are visible in the soil, the crop and the final product. But sustainability does not always work that way. Global buyers increasingly ask for cleaner food, cleaner fuel, lower-carbon supply chains, better traceability and stronger environmental practices. The demand is there, but the willingness to pay is often less clear.

Brazil sits right in the middle of this paradox. The country has built real advantages in renewable energy, ethanol, biodiesel, no-till farming, biological inputs, integrated crop-livestock systems and preserved native vegetation on private land. It is not a perfect story, and the challenges around deforestation, land use and enforcement are real. But there are also important moves in the right direction.

That was the sharper point behind a recent CNN interview with Brazilian agricultural analyst Marcos Fava Neves. Brazil already has many of the things global markets say they want. The question is no longer only whether Brazil can produce cleaner food, cleaner fuel and lower-carbon supply chains. The question is who will pay for them.

That is where the green gap sits. Brazil does not lack green initiatives. It lacks the business models to connect green value in the field with buyers, investors and lenders who say they want it. Corn ethanol, biological inputs and renewable-powered data centers all point to the same challenge: Brazil has the assets, but still needs better ways to capture the value.

Beyond the Bushel

Soybeans, corn, sugar, coffee, cotton and beef have turned Brazil into one of the world’s top agricultural powerhouses. But producing volume and capturing value are not always the same thing.

The recent move by Amaggi, one of Brazil’s largest grain and agricultural groups, to buy 40% of FS, a major corn ethanol producer in Mato Grosso, shows how that may be changing. Brazil’s agricultural giants are starting to look beyond the bushel, turning crops into fuel, feed, energy and potentially lower-carbon products closer to where they are produced.

That’s important in a state like Mato Grosso, where distance is always part of the business model. Even an efficient farmer can see margins squeezed by freight, storage and port bottlenecks. Local processing does not solve every problem, but it gives the system another outlet before logistics costs eat too much of the upside.

Corn ethanol fits neatly into Brazil’s broader green stack. It supports biofuel supply, creates byproducts for animal protein, uses biomass and can open the door to carbon-linked opportunities if emissions are properly measured and rewarded. For Brazil, the bigger prize is keeping more value close to where the crop is grown.

Micro Margins

Brazil is already one of the world’s largest and fastest-growing markets for biological inputs, and this growth is much more than a niche sustainability trend. It reflects a practical shift in how farmers are managing cost, risk and resilience in tropical agriculture. Many traditional inputs are imported, which exposes farmers to currency swings, supply shocks and high prices, while many products were originally developed for Northern Hemisphere agriculture, where soils, seasons and pest pressure are very different. What works in Iowa or France does not always work the same way in Mato Grosso or Bahia.

Biological inputs offer a more local and adaptive path because they use living organisms, natural compounds or biological processes to support crops and soil health, instead of relying only on fossil-based chemistry. When they work well, they can help reduce chemical dependency, improve soil biology, manage pests and diseases, and make production more resilient. For farmers, the appeal is practical rather than ideological. Lower cost, healthier soil and fewer surprises in the field usually beat a nice ESG slide.

This is also where AI could become useful, especially because biologicals are complex and highly dependent on local conditions. Performance can vary according to the microbe, the crop, the soil, the weather, the timing and the farm practice around it. Better data and AI tools can help identify promising strains, match products to local conditions and reduce some of the trial-and-error that still slows adoption. The opportunity is not only in discovering better products, but in helping farmers and agronomists understand where, when and how to use them.

The missing piece is still the business model. If a farmer improves soil biology, reduces chemical use or builds more resilient production, there should be a clearer path to economic recognition through cheaper credit, better insurance, stronger contracts or preferred access to buyers. Biologicals show that greener tools can scale when they solve real farm problems. The next step is building the credit, data, insurance and buyer models that reward those choices more directly.

Cleaner Data

Brazil’s renewable energy advantage is starting to create opportunities beyond farms, mills and fuel plants, especially as solar, wind and biomass generation expand faster than parts of the grid can absorb. Sugarcane mills are a good reminder that Brazil’s green stack is already more connected than many outsiders realize. They do not only produce sugar and ethanol; they also generate electricity from bagasse, the fibrous residue left after crushing cane, and can sell excess power back into the grid.

Minter, a mobile data center startup, is tackling a similar value-capture problem in renewable power by installing container-based units directly at generation sites. The units process data close to where electricity is produced, helping power producers monetize energy that might otherwise be curtailed when the grid cannot absorb supply. The company already operates a 20 MW site in Xique-Xique, Bahia, and plans to reach 40 MW by the end of 2026 and 500 MW by 2029.

The model follows the same logic as corn ethanol and biological inputs. It takes an existing green asset and finds a better way to monetize it. In a country with growing renewable generation, transmission bottlenecks and rising demand for computing power, that kind of business model can turn stranded or underused energy into a new revenue stream.

Larger data center projects, including Omnia’s renewable-powered project in Ceará, point in the same direction at a different scale. Brazil’s clean energy base can attract new industries, but the more important question is what gets built around them. For agriculture, better data can help prove and price what is happening in the field, from traceability and carbon intensity to biological input performance and credit risk.

Green Gains

The problem is not that farmers and agribusinesses need another lecture on sustainability. Most already understand where the pressure is coming from: buyers want cleaner supply chains, banks want lower risk, regulators want more data, and consumers want reassurance that the food and fuel they buy are not creating hidden environmental costs. The harder question is how those expectations become a better price, a better contract or a lower cost of capital for the people doing the work.

Better business models will have to connect what happens in the field with how value is shared across the chain. A farmer who improves soil health, uses biological inputs, reduces chemical dependency or lowers carbon intensity should have a clearer path to cheaper credit, better insurance, preferred supplier status or longer-term offtake. A biofuel producer with a lower-carbon product should be able to capture that difference through the fuel market. A renewable energy producer with stranded power should have more ways to turn that electricity into revenue.

This is where technology becomes useful, but only if it stays close to the money. Traceability, satellite monitoring, carbon accounting, farm data, AI tools and digital credit models all sound impressive, but they only matter if they help reduce risk, prove performance, open a market or support payment. Otherwise, they become another layer of reporting that farmers are asked to carry without seeing the benefit.

Brazil has plenty of green activity already. What it needs now are more bridges between practice and payment. Corn ethanol, biological inputs and cleaner data all show different versions of the same opportunity: take value that already exists in the system, prove it better, process it smarter and make sure more of the upside reaches the people and companies creating it.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Hungry Capital

AI has already swallowed the venture capital agenda. According to Crunchbase, AI funding reached $211 billion in 2025, up 85% from $114 billion in 2024, representing roughly half of all global venture funding. OECD data puts the figure even higher, estimating that AI firms captured 61% of global VC investment in 2025, equal to $258.7 billion out of $427.1 billion.

This huge wave of investment was needed to build models, infrastructure and computing power before AI can be applied widely across the economy. But as this layer becomes more consolidated and harder for new entrants to win, venture capital will start looking for other places where AI can create real business value.

The next wave of AI investing will flow to where models can be applied to large markets, complex workflows and expensive problems. Legal, healthcare, finance and logistics are already attracting attention. Agriculture will also be part of that search, and this is where Brazil comes in.

Brazilian agriculture has scale, complexity, data, pain points and global relevance. It is also full of workflows that are manual, fragmented, relationship-driven and only partially digitized. The same complexity that made agriculture hard for traditional software may be what makes it so attractive for AI.

Big Field

Agriculture is often described as one sector, but that misses the point. It is really a dense network of decisions that stretch from the soil to the port, from the farm gate to the balance sheet, and from weather risk to working capital.

A farmer is not only deciding what to plant. He is managing weather risk, input costs and market timing. A cooperative is not only selling inputs. It is advising producers, extending credit and coordinating logistics. A grain trader is not only buying soybeans. It is managing freight, price risk and counterparty exposure.

Vertical AI becomes interesting in agriculture because so many of these decisions already shape margins, risk and cash flow. Selling grain, approving credit and spotting disease pressure are all high-value areas where better intelligence can make a measurable difference.

Brazil gives this thesis unusual scale. Agribusiness exports reached a record $169.2 billion in 2025, accounting for 48.5% of Brazil’s total exports. Brazilian agriculture is not a small market waiting to be discovered by technology investors. It is already one of the country’s core economic engines and one of the most important food and agricultural systems in the world.

Messy Moats

Every investor likes the word moat. In agriculture, the moat may come from the mess.

Brazil’s ag economy still runs through a blend of WhatsApp messages, field visits, invoices, satellite images, phone calls, spreadsheets, informal knowledge and long-term relationships. For traditional software, that can be difficult to organize. For AI, it is raw material.

To work in a vertical like agriculture, AI needs more than a strong model. It needs industry context, proprietary data and access to processes that outsiders cannot easily understand or copy. Agriculture has all three, although they rarely come neatly packaged. Information is often scattered across people, machines, fields, buyers, lenders and service providers, which makes the problem harder but the opportunity more defensible.

Moving grain in Brazil, for example, is a seasonal, regional and highly volatile decision system shaped by harvest timing, rail capacity and port lineups. A platform that understands those variables is doing more than digitizing freight. It is building intelligence around one of the biggest cost centers in the value chain.

The same idea applies across animal protein, biological inputs and crop monitoring. The companies that can capture and interpret signals from these activities will not only have better software. They will have a stronger position in the value chain.

Agrifoodtech funding has been through a painful reset globally, but investor interest is moving toward technologies that can create real defensibility, including AI, robotics, biotech and advanced sensing. That should favor companies solving hard problems, not those adding a thin AI layer to old software.

Credit Gateway

The easiest bridge between AI capital and agriculture may be the fintech route.

Latin America’s venture story has already been shaped by fintech. Many of the region’s most successful startups used better data, better interfaces and better underwriting to reach customers that traditional financial institutions ignored, misunderstood or priced poorly.

Agriculture has a similar opportunity, but with more variables and muddier boots. In consumer fintech, credit risk depends on income, spending patterns and repayment history. In agriculture, repayment also depends on production, weather and commodity prices. A farmer may be a good borrower and still have a bad year because yields fall, prices move or freight costs eat into margins.

That makes ag credit harder to model, and more valuable when done well. Traditional credit models are often too blunt for agriculture, while AI can process richer signals from satellite imagery, production history, invoices, market prices and logistics data.

Agfintech is already one of the most venture-compatible categories in agriculture because it has clear pain, large markets, recurring demand and business models investors understand. In The Yield Lab’s Latam portfolio, around 40% is already made up of agfintech companies. Much of the other 60% could eventually offer some form of embedded finance, because many already have customer relationships, operational visibility and rich data.

That may be the real unlock. Credit does not need to remain a standalone product. A marketplace can offer credit, a logistics platform can finance freight, and a sustainability platform can support transition finance. The first wave of agfintech digitized access to capital. The next wave could use AI to predict risk, price capital and personalize finance around the real economics of the farm.

Seed Scale

If global AI investors start looking seriously at Brazilian agriculture, they will not arrive searching for ideas. They will be looking for companies with traction, proprietary data, scalable distribution and enough maturity to absorb large Series A or Series B rounds.

That creates both an opportunity and a risk. The opportunity is that Brazilian agriculture fits what vertical AI capital will likely want next: large problems, complex workflows, valuable data and real economic pain. The risk is that many AI-native agriculture startups may still be too early when that capital arrives.

This is why early-stage conviction matters. Brazil needs more founders building AI-native solutions for large, scalable problems in agriculture. It needs more data-driven pilots with farmers, corporates, cooperatives and financial institutions. It also needs more events, demos and practical use cases that move the market from general AI excitement to measurable business value.

That is exactly the kind of momentum we hope to build at our next AI in Ag event on May 28 in São Paulo, with participation available both in person and online for Portuguese speakers. The goal is to showcase real use cases, connect the right people and help move AI in agriculture from potential to practice.

When capital moves from models to markets, Brazilian agriculture should be ready to serve.

Thanks for reading.
KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Digital Dream

Agrishow, Brazil’s largest agricultural machinery fair, took place this week in Ribeirão Preto, about four hours from São Paulo. For one week, more than 200,000 people descend on the city to look at tractors, planters, sprayers, combines, drones, and whatever else promises to make farming a little more productive or a little less painful.

If you have never been, it is hard to describe the scale. There are machines everywhere. Huge machines. Machines that look more like mining equipment than farm tools. For decades, that was the point. Agrishow was where Brazilian agriculture came to show its muscle.

I was there this week for two panels, one on AI in agriculture and another on the evolution of the AgTech ecosystem. Machinery still dominated the landscape. Farmers still need to plant, spray, harvest, haul, store, and sell. No app is going to pull a planter through heavy soil.

The next layer of value is forming around the machine. It is in the software behind it, the data coming off the field, the image from the drone, the weather alert, the price signal, the WhatsApp message from the agronomist, and the ability to turn all of that into a better decision before the window closes.

That is the promise. A farmer should be able to ask a simple question during the day and get a useful answer. Should I spray now or wait? Is this freight quote fair? Should I sell soy today? Is this pest pressure showing up elsewhere? Can I afford to test this biological input? What machine should go where first?

AI makes that kind of tool much easier to imagine and much cheaper to build. The problem is that the tool is only as good as what it knows. On most farms, a lot of what matters still sits outside a clean database. It is in photos, voice notes, field visits, invoices, machine operators’ comments, WhatsApp groups, and the farmer’s own instinct.

That is where the next challenge begins.

Dating Game

In the past, some startups have been guilty of treating farmers like unpaid data-entry employees.

Fill out this form. Upload this file. Tag this field. Enter this activity. Update this dashboard. Log the spray. Record the cost. Connect the machine. Come back next week and do it all again.

That might look fine in a product demo. In the field, it dies quickly. It is also one of the main reasons many AgTech startups struggled to understand why farmers were not engaging, even when the product made sense on paper.

Farmers are already juggling weather, people, machinery, inputs, credit, prices, pests, logistics, and risk. Asking them to feed another system without giving something useful back straight away is a hard sell. Agriculture has enough friction. It does not need software adding homework.

That is why data capture is really a relationship problem.

You do not get the farmer’s information on the first date. You have to earn it.

The exchange has to be simple. The farmer gives something small and natural, like a photo, voice note, WhatsApp question, freight quote, planting update, pest observation, input price, or yield number. In return, he gets something useful quickly: a benchmark, alert, comparison, recommendation, market signal, risk warning, or next-best action.

This give-to-get exchange matters because farmers like comparison. They want to know what is happening over the fence. Am I paying too much for inputs? Is my freight quote high? Are neighbors seeing the same pest pressure? Is my yield on track? Should I sell now or wait? How does my cost structure compare with similar farms?

That was part of the early power of Farmers Business Network in the United States. Farmers shared information because they received benchmarks and transparency back. The comparison created value, and the value encouraged more sharing.

Brazil may be even more suited to this model because farmers already live on WhatsApp. It is the unofficial operating system of the countryside. Prices, photos, invoices, field problems, freight quotes, agronomic doubts, worker updates, and commercial negotiations already flow through it every day.

The best data capture strategy may look less like data capture and more like a useful conversation.

A farmer sends a photo of a leaf disease and receives a likely diagnosis, plus an alert that similar cases are appearing nearby. Another shares a freight quote and learns whether it is above or below the regional range. Another asks whether to sell soybeans today and receives local price intelligence, storage considerations, and cash-flow context.

Every useful response builds trust. Every trusted interaction creates more data.

That is how the farm data layer starts to form. Through value, not forms.

The App Trap

AI has made it easier than ever to build the front end of farm intelligence.

A startup can now create an app, agent, chatbot, dashboard, or recommendation engine much faster and cheaper than before. That is a good thing. It lowers the cost of experimentation and opens the door for more founders to build for agriculture.

It also creates a trap.

Build the app faster than the data relationship underneath it, and the product may look intelligent without being useful. A smart-looking recommendation sitting on incomplete data is still just a guess in a nice interface.

And AI language is everywhere now. Every platform wants to become a copilot. Every dashboard wants to become an agent. Every farm management tool wants to offer recommendations. That may be the right direction, provided the system knows enough about the farm to recommend something practical.

A satellite image can show crop stress, though not always why it happened. A drone can identify weeds, though not whether the sprayer is available. A sensor can measure soil moisture, though not whether the farmer has the labor, cash, or input stock to act. An ERP can show inventory, though not what the agronomist saw in the field yesterday.

A chatbot can sound clever. If it is disconnected from the farm’s real operating context, it is still mostly guessing.

This is the difference between digital decoration and decision infrastructure.

The app is not the intelligence. The agent is not the intelligence. The dashboard is not the intelligence.

The intelligence comes from knowing what is actually happening, connecting it with context, and helping someone act in time.

Join the Dots

The end goal is not a better dashboard. It is an easy-to-use tool that helps the farmer make better decisions throughout the day, across whatever activity matters at that moment.

That decision may be agronomic. Should I spray? Should I irrigate? Should I replant? Is this pest pressure serious or still manageable?

It may be operational. Which machine should go where first? Do I have enough labor? Is the sprayer free? Did the input arrive? Will rain stop the operation?

It may be financial. Should I take credit now? Should I delay a purchase? Can I afford a biological input trial? Does this technology pay back this season, or only over three years?

It may be commercial. Should I sell grain today? Should I store? Is this basis attractive? Is the freight quote too high? Should I buy inputs now or wait?

It may involve machinery, livestock, land, sustainability, insurance, certification, or market access.

One data point rarely makes a farm decision.

A field photo matters. A weather forecast matters. A satellite image matters. A machinery alert matters. A grain price matters. A cash-flow update matters. None of them are enough alone.

The farmer needs a system that can connect agronomy, weather, machinery, labor, finance, logistics, input prices, market signals, risk, and sustainability into one practical decision layer. The more relevant information the system can access, the better it can help the farmer make decisions that are efficient, profitable, and sustainable.

This is where AI becomes powerful. It can connect more variables than any human can hold in their head at once. It can notice patterns, compare scenarios, structure messy information, and suggest next steps.

Agriculture’s challenge is that too many of those variables are still scattered across systems, people, phones, machines, documents, and memory.

AI is very good at connecting dots. Agriculture’s problem is that too many of the dots are still sitting in people’s heads, phones, notebooks, and WhatsApp groups.

The real breakthrough will come when those dots can be captured naturally, connected quickly, and returned to the farmer as useful guidance.

Missing Link

The next phase of farm AI will not be won by the company with the prettiest dashboard or the loudest AI claim. It will be won by whoever solves the link between farmer behavior, low-friction data capture, and useful decisions.

That link has three parts.

First, the farmer must be willing to engage. The interaction has to feel natural. WhatsApp, voice, photos, videos, and simple questions will often beat another app login. The system needs to fit the farmer’s workflow, not force the farmer to fit the system.

Second, the farmer must get value back quickly. A benchmark. A warning. A comparison. A recommendation. A market signal. A better price. A lower risk. The exchange has to be clear. Give information, get insight.

Third, the information has to be connected. A farm is not a spreadsheet with soil attached. It is a living system of biology, machinery, finance, labor, weather, logistics, markets, and instinct. Capturing individual pieces is useful. Connecting them fast enough to change decisions is where the value lives.

Agrishow showed plenty of digital dreams this week. Some will be real. Some will be premature. The test is simple: does the product help the farm understand itself better, and does it turn that understanding into a decision?

The first wave of digital agriculture tried to make farmers use software. The next wave of AI will need to make software understand farmers.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Hot and Cold

In my last article, I mentioned how impressed I was by the sophistication of the UK AgTech ecosystem. Lots of funding, strong talent, and serious technology. What it’s missing, however, is a market large enough to help these startups scale to venture capital heights.

A lot of the solutions I saw were agronomic in nature and aimed directly at farmers. That is a very tough market. It’s fragmented, seasonal, and slow to build. Most farmers are busy running the farm, and many are not looking to spend extra time or money trying out the latest technology unless the value is very clear.

That creates a problem for venture-backed startups. You can build a good product, raise serious money, and still find yourself selling into a market that is hard to grow quickly. In most cases, the technology is not the main constraint. The real issue is that the home market may simply be too small or fragmented.

This is where Brazil starts to look attractive. The market is larger, runs year-round, and the problems are often bigger and more urgent, including many outside the farmgate that are B2B in nature and easier to scale. The opportunity is there, even if getting in is not always straightforward. The three examples that follow show different ways that fit can work in practice.

Bounty Hunt

Building in the UK does not always mean building for the UK. In some cases, the much bigger opportunity is global. One good example of this is Tropic, a startup that came out of the Norwich Research Park and now operates internationally.

Tropic is a biotech company focused on improving tropical crops through gene editing. It has raised more than $190 million and is already operating well beyond the UK. Its work in bananas shows why.

The global banana industry still depends heavily on a single variety, Cavendish, which makes it unusually vulnerable. Disease pressure is high, and Tropic highlighted that around 50% of bananas are lost along the supply chain. One of the biggest threats is Panama disease, a soil fungus that can stay in the ground for up to 50 years. That is why resistant varieties matter so much.

Tropic is working on traits linked to disease resistance, but also on a gene to stop oxidization aimed at reducing browning and helping the fruit last longer after harvest. The company already operates in Brazil and other parts of Latin America, as well as India, the Philippines, and the Middle East.

Remote Edge

If Tropic shows how UK science can be built around global crop problems, Lacuna Space shows how a solution developed in the UK can become much more relevant in a market like Brazil.

Lacuna is a satellite company focused on connecting low-power sensors in remote areas. It is not trying to compete with Starlink or provide broadband internet. Its model is much simpler: help small field devices send small amounts of data from places where normal connectivity does not reach.

That fits Brazil well. A farm may have internet at the office, but that does not mean the field is connected. And in many cases, the sensor only matters if it is out in the field.

One reason Lacuna has landed well in Brazil is that it has found trusted partners such as Embrapa, USP, and Corteva to work on an important problem: black sigatoka in bananas. The system combines sensor data with weather, satellite imagery, and AI to improve spray timing and reduce fungicide use.

Brazil has the size, the crop diversity, and the rural connectivity gaps to make a product like this much more relevant than it might be in its home market.

Crop Pick

If Lacuna is about getting data out of the field, Fruitcast is more about what to do with that data. It also points to another useful strategy for startups: pick a high-value crop, where the pain is clearer, the margins are higher, and the return on a better tool is easier to prove.

Fruitcast started with a simple problem in strawberries. Growers often do not know with enough confidence how much fruit is actually in the field, which creates problems for labor planning, contracts, waste, and margins. In a high-value crop, a weak forecast quickly becomes an expensive one.

The company’s solution is quite practical. Someone walks the crop with a phone or camera, uploads the video, and Fruitcast uses computer vision to count fruit and flowers, estimate maturity, and project yield. It then adds weather, historic yield data, crop cycles, and variety information to build a digital view of the farm.

This matters because better forecasting gives growers more time to plan labor, manage supply, and deal with buyers if volumes are likely to be above or below plan. In a crop like strawberries, those gains can show up quickly, which makes it a much easier place for a startup to prove value than in broadacre.

Many of those high-value crop opportunities sit outside the UK today, but that may also change over time. Dyson Farming’s strawberry operation showed how far that can go when the crop is valuable enough to justify more technology and more capital. Using large greenhouses, novel varieties, and a hybrid vertical growing system, Dyson is producing strawberries all year round and has managed to triple yields from the same area.

Hard Ground

While Dyson hints at what the future can look like, all-year-round farming, especially in high-value crops, Brazil already offers that now. So why have more UK and other Northern Hemisphere startups not made the move? Because Brazil is still a hard nut to crack.

A company can have strong technology and still struggle if it does not have the right local partner, the right first customer, the right crop focus, or the right route to market. Brazil is not a place where you just land, translate the deck, and hope things work out.

That is especially true in agriculture. Conditions vary a lot by region, crop, and business model. What works in one market may need real adaptation in another. In some cases, the product fits well, but the go-to-market does not. In others, the issue is pricing, local support, or simply finding the right place to start.

Brazil may offer the scale many foreign AgTech startups are missing at home. But scale on its own is not enough. The winners will be the ones that find the right entry point, the right partners, and the patience to adapt.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Market Lens

Having grown up on a farm in Ireland, I’m fairly familiar with the European AgriTech scene, so I was not expecting this one-week ARISE* trip across the UK ecosystem to show me anything new. I returned to Brazil, however, more surprised and impressed than I had expected.

What stood out was not just the science, the quality of the centres, or the funding behind them. These institutions seem to have matured beyond a pure research and technology lens and are taking a harder look at market opportunities, including ones farther afield.

The UK has many of the ingredients needed to build strong AgriTech: talent, public funding, serious research infrastructure, and institutions like Norwich Research Park, the John Innes Centre, The Sainsbury Laboratory, the Lincoln Institute for Agri-Food Technology, Cambridge-based Niab, and Warwick Manufacturing Group. What it does not have is a large addressable market. British agriculture is relatively small, fragmented, and seasonal, which limits its attractiveness for venture capital.

That reality seems to be changing the way some of these centres think. The focus is widening from developing the science to finding the right market, the right test bed, and the right commercial path. Brazil kept coming into the picture for exactly that reason.

Southern Appeal

Norwich Research Park made that shift clear. The scale is impressive, but the more important point is that this is no longer just a cluster of strong research institutions. It is becoming a platform that is thinking more seriously about where science can travel and where it can find the biggest commercial opportunity.

According to Jonathan Clarke of John Innes Enterprise, the traditional model often centred on attracting foreign startups and talent into the UK through funded projects and research collaborations. That model is now widening. The UK has strong science, serious funding through bodies like the Biotechnology and Biological Sciences Research Council (BBSRC) and UK Research and Innovation (UKRI), and deep institutional capability, but it also needs larger and more commercially relevant markets if that science is going to stretch beyond its home market.

Brazil offers exactly that kind of market. The UK has excellent research capability, but a smaller, fragmented, and seasonal agricultural market. Brazil offers larger scale, year-round production, tropical crops, climate pressure, and a much bigger commercial opportunity in both agriculture and climate. For UK AgriTech looking to move from lab to land, some of that land will need to be outside the UK.

According to James Canham of GetGenome, a non-profit spinout from The Sainsbury Laboratory, around 95% of genomes in the main databases come from the Global North. That points to a major blind spot in global science and a major opportunity as well. If the data is concentrated in the North, then a large part of the future value may come from helping researchers in the Global South build better genomic infrastructure of their own.

Farm Reality

If Norwich was strong on market direction, the Lincoln Institute for Agri-Food Technology (LIAT) was strong on practical reality. The conversation there was less about abstract potential and more about what it takes to make a technology survive in an agricultural setting.

One example was the effort to make a robot more farm-ready by giving it extra torque. That is not a flashy detail, but it tells you a lot. Agriculture is a rough operating environment, and machines need to cope with mud, weight, uneven ground, and long working days. Design choices like that show a centre thinking about the job, not just the demonstration.

The idea of developing stacks that can sit on top of different robotic platforms was also interesting. That feels like a more practical route to scale than tying everything to one machine. In AgriTech, the strongest solution is often the one that can be adapted, reused, and deployed across different contexts without starting from zero each time.

I also liked LIAT’s “pull, push, foundry” framework for identifying market opportunity. Some technologies are pulled by market demand. Others are pushed by research. Others need a more collaborative foundry model where industry, researchers, and builders work through the problem together. Whatever the starting point, the lesson was the same: the technology has to earn its way onto the farm.

Scale Matters

Warwick added another layer to the story. If Lincoln was about field readiness, Warwick was about manufacturing readiness. That may sound less exciting, but it is where many AgriTech ideas either become real businesses or quietly stall.

Warwick Manufacturing Group at the University of Warwick seemed to approach the problem from exactly that angle. The real question is not just whether a technology works. It is whether it can be produced, repeated, integrated, and scaled in a way that makes commercial sense.

What stood out there was the multidisciplinary mindset. Life science, software, hardware, engineering, and business are all part of the same conversation. That matters because AgriTech tends to fail where those disciplines do not connect. A promising product can easily get stuck between technical proof, operational fit, and manufacturing economics.

That manufacturing lens also feels like a sign of maturity. Too much AgriTech still gets judged too early by whether it is novel or technically impressive. Warwick seemed more interested in whether it could survive contact with reality. In that sense, the move from lab to land also depends on a move from prototype to production.

Support Rails

What also stood out across the trip was that the centres are not operating in isolation. Around them sits a layer of programmes, partnerships, and support structures that help move ideas forward. That part matters because strong science on its own is rarely enough.

According to Gareth Arnott, Head of Biological Sciences at Queen’s University Belfast, the Momentum One Zero programme is a good example of this. Instead of funding research and worrying about the market later, it is building a more applied framework around digital innovation, data, and commercial collaboration. That kind of support can be especially useful in agriculture, where the path from science to business is rarely straightforward.

The same applies to the wider discussions around gateway programmes, soft-landing models, and institutions like Barclays Eagle Labs. The UK seems to be getting more serious about the infrastructure around innovation, not just the invention itself. That includes helping startups and technologies find the right networks, markets, and partners beyond their home base.

That may end up being one of the most important lessons from the trip. AgriTech does not just need better science. It needs better rails around the science so good ideas can move more smoothly from research centre to company, from company to pilot, and from pilot to market.

*ARISE (Agri-Tech for Resilience, Innovation & Sustainable Ecosystems) is a UK government-funded Tactical Fund project led by Agri-TechE in partnership with AgriTIERRA, Earthbase, Embrapa, and The Yield Lab Institute. The trilateral knowledge exchange program across the UK, US, and Brazil brought together participants from Niab, Newcastle University, Harper Adams University, the Donald Danforth Plant Science Center, the Taylor Geospatial Institute, Washington University in St. Louis, CREA and Universidad Austral in Argentina, AgCenter Colombia, Tec de Monterrey in Mexico, AgTech Chile, APTA Hub in Brazil, and The Yield Lab Latam.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Living Code

Agriculture has always depended on biology. Soil microbes cycle nutrients, insects pollinate crops and fermentation organisms transform plants into fuel, food and industrial products. For most of farming history these processes were simply part of the background conditions farmers learned to work with.

Technology has traditionally focused on improving how we observe those systems. Satellites monitor crop stress, sensors track machinery performance and artificial intelligence models help predict disease outbreaks or yield patterns. These tools generate better information, but they mostly sit outside the biological processes that actually drive productivity.

What is beginning to change is that technology is moving closer to those living systems themselves. Instead of simply measuring biology, new approaches are starting to influence how microbes behave, how fermentation organisms perform and even how insects interact with crops. In effect, parts of agriculture’s biological infrastructure are becoming easier to influence and manage.

Bee Business

Pollination is one of the clearest examples of this shift. Many crops depend on insects to move pollen between flowers, yet farmers historically had little control over how efficiently that process occurred in the field. Beekeepers could place hives nearby, but directing pollination activity itself remained largely unpredictable.

Companies like Beeflow are beginning to change that dynamic. Using scent conditioning techniques, the company trains bees to prioritize specific crops during flowering periods, increasing the probability that pollinators concentrate on target plants instead of surrounding vegetation. The result can be higher pollination rates and improved yields in crops that rely heavily on insect activity.

The idea is simple, but it changes the level of control farmers have in the field. Pollination remains a natural process, yet technology is beginning to guide it in measurable ways. With tools like this, insects start to look less like passive ecosystem actors and more like biological tools that can improve agricultural productivity.

Microbe Mechanics

A similar transformation is happening at the microbial level. Brazilian biotech startup Bioinfood, for example, is developing yeast strains designed to improve ethanol production efficiency in sugarcane and corn processing plants. Small improvements in microbial performance can increase output and reduce losses without requiring major changes to industrial infrastructure.

Microorganisms are also becoming a core component of crop management. Biofertilizers, biopesticides and biostimulants use bacteria, fungi and plant-derived compounds to improve soil health, nutrient uptake and crop resilience. These products increasingly complement or replace synthetic fertilizers and pesticides that have dominated agriculture for decades.

Adoption is already significant in Brazil. More than half of Brazilian farmers use at least one biological input, and bioinputs are applied on roughly 26 percent of the country’s planted area, placing Brazil among the most advanced biological input markets in the world. In crops such as soybeans, inoculants are used across the vast majority of hectares, reflecting decades of research and farmer familiarity with microbial solutions.

Brazil Lab

Brazil’s agricultural system makes it a natural environment to test these technologies at scale. The country combines vast crop production with extraordinary biodiversity and one of the world’s strongest agricultural research ecosystems, led by institutions such as Embrapa and major universities.

Tropical agriculture also creates the conditions where biological solutions are most relevant. Crops grow year round, pest and disease pressure is constant, and farmers operate across a wide range of climates, soils and production systems. That complexity forces technologies to perform under real-world conditions rather than controlled environments.

The economic opportunity is also expanding quickly. Bioinputs still represent only a small share of Brazil’s roughly US$19 billion agricultural input market, leaving substantial room for substitution as farmers look for more resilient production systems. If adoption continues to accelerate, the domestic bioinputs market alone could exceed US$8.5 billion by 2035.

Biological Stack

Agriculture has evolved through several major technological waves. Mechanization expanded scale. Chemical inputs dramatically increased productivity. Digital tools improved how farmers measure and manage increasingly complex operations.

Biological technologies are now joining that stack. Pollination systems that guide insects, microbial inputs that reshape soil ecosystems and fermentation organisms optimized for industrial production are all examples of tools working directly within living processes rather than simply monitoring them.

The idea that biology can behave like software is no longer purely theoretical. Researchers recently simulated an entire living cell on a computer, modeling genetic processes, metabolism and cell division across a complete life cycle. While that work sits at the frontier of computational biology, it signals how quickly our ability to understand and influence living systems is advancing.

Brazil may end up sitting at the center of this transition. With large-scale agriculture, strong research institutions and a rapidly expanding biological inputs sector, the country is becoming one of the most important environments in the world for testing how living systems can improve the productivity, resilience and sustainability of global food production.

Thanks for reading.
KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Smart Fixes

Last week I wrote about large companies embedding artificial intelligence into their operating systems. This week I’m shifting the lens. Startups tend to enter where the system is already under strain. They focus on the bottlenecks farmers experience every season but often accept as part of the job.

Some of those strain points sit in the soil itself. Others surface when it is time to sell grain. And some appear when trucks, rail and ports struggle to keep pace during harvest. In each case, the challenge is rarely a lack of data. It is the cost, delay or complexity involved in turning that data into efficient decision making.

What stood out at our event in Sao Paulo last week (Inteligent Ag: AI in Action) was how specific these solutions are. These founders are not trying to automate entire farms. They are fixing narrow loops where uncertainty quietly erodes profits. Soil diagnostics that take too long. Grain prices negotiated without the full picture. Logistics systems that creak when volumes spike.

Below are three examples of AI being applied at those pressure points, starting at ground level.

Microscopic Margins

Nematodes are microscopic worms that live in the soil. Some are beneficial. Others attack plant roots, reducing nutrient uptake and lowering yields. In Brazil alone, plant-parasitic nematodes are estimated to cause roughly $7 billion in crop losses per harvest, with soybeans accounting for more than $3 billion of that total. Yet only about half of producers actively monitor the problem.

Diagnosis has traditionally relied on manual microscopy. Technicians examine samples under standard magnification, classify organisms visually and group them broadly by genus. Species-level identification often requires additional testing. The process is slow, dependent on specialist expertise and often inconsistent across labs. Different nematode species require different management strategies. When classification lacks precision, farmers sometimes apply the wrong product, too much, or too little.

Veridi, a Dutch biotech startup that recently launched in Brazil, built a robotic microscope specifically for nematode analysis. The system captures ultra-high-resolution images and processes them through computer vision models trained to distinguish species with more than 95 percent accuracy (compared with around 40% the traditional manual way).

Beyond identifying harmful species, the platform also classifies free-living nematodes, which serve as biological indicators of soil health, carbon cycling and regenerative practices. In Europe, soil monitoring regulations now incorporate nematode-based metrics. In Brazil, this layer of analysis remains largely untapped.

Veridi does not sell the equipment. Instead, it installs the system at partner laboratories at no upfront cost and charges per analysis, typically below the price of traditional nematode diagnostics. Rather than building a new lab network, it upgrades existing infrastructure. The result is faster turnaround, greater consistency and more precise soil intelligence.

Commercial Edge

Brazilian farmers are among the most efficient producers in the world. They manage tropical biology, double cropping and tight planting windows with impressive precision. Commercialization, however, is more complicated.

Farmers here must navigate currency volatility, regional basis spreads against Chicago, uneven logistics and limited hedging tools. Compared with U.S. farmers operating in deeper financial markets, Brazilian producers face more variables at the moment of sale. In volatile markets, even a short delay can mean missing a pricing window.

Grão Direto, Brazil’s first digital grains marketplace, facilitated more than 12 million tons of grain transactions last year. That scale produces something rare in Brazil’s physical market: a large dataset of executed trades. Quoted prices are one thing. Real completed transactions provide a clearer signal of where business is actually being done.

Building on that data, the company developed Airton, an AI-powered commercial assistant that operates inside WhatsApp, the platform nearly every Brazilian farmer already uses daily. Producers can ask about soybean prices, regional basis levels, currency movements or market context in plain language and receive responses grounded in real transaction data.

Airton does not replace the commercial relationship. It shortens response time and expands access to information. Producers can receive price alerts, track movements during the trading day and formalize transactions through integrated trading partners.

Long Haul

Brazil is continental in scale. Grain is grown from the southern most state of Rio Grande do Sul up to Roraima in the northern hemisphere. Moving crops across that geography isn’t a minor logistical issue — it’s one of the biggest challenges Brazilian agriculture faces.

During peak harvest, that constraint becomes visible. Trucks tighten. Freight rates rise. Inputs for the second crop need to move toward farms while grain moves toward ports. Delays compound quickly when millions of tons are in motion across thousands of kilometers.

Brazilian startup goFlux operates at the center of that movement. Through its digital freight marketplace, the company transacts tens of millions of tons of agricultural cargo each year, generating a live dataset on how freight behaves across routes, regions and seasons.

Using its Intelligence platform, goFlux recently modeled 3.6 million logistics scenarios for a large ethanol producer weighing a long-term decision: remain exposed to road freight volatility or commit to rail capacity for the next 15 years.

Rail required upfront investment and long-term contracts. Road offered flexibility but rising structural pressure as ethanol production expands and competition for tanker trucks increases. Freight cannot be hedged like currency or Chicago prices. Based on the modeling, the company chose to lock in rail positioning to reduce exposure and protect margin.

Sharper Edges

Put these three cases side by side and you start to see the common thread. None of these companies are trying to reinvent agriculture from scratch. They are working on the edges — the places where friction shows up every season.

In the soil, that friction looks like slow diagnostics and imprecise treatment. At the moment of sale, it shows up as fragmented information and missed price windows. In logistics, it appears when harvest volumes overwhelm roads and long-term freight exposure goes unmanaged. Farmers and operators have learned to live with these constraints. Startups are choosing not to.

Brazilian agriculture already runs at scale. What these companies are doing is tightening the parts that leak value — a lab result delivered faster, a price confirmed with more confidence, a freight decision made before the squeeze hits. None of it is dramatic on its own. Together, it makes the system more resilient and a little more profitable for farmers.

Veridi, Grão Direto and goFlux presented these cases at The Yield Lab Latam AI in Agro event (26/2) in São Paulo, in partnership with the Brazilian Rural Society (SRB). The event was held in Portuguese. You can watch the full recording here

Thanks for reading.
KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Heavyweight Thinkers

Most conversations about AI in agriculture focus on startups building new tools or experimenting in controlled settings. That is usually where innovation begins. Smaller teams can move quickly, adjust fast, and experiment without too much internal resistance.

What is less visible is how large agribusinesses are beginning to incorporate AI into their day-to-day operations. These are organizations managing supply chains, machinery fleets, animal production systems, and agronomic programs across thousands of hectares. Any new technology has to work within existing processes, compliance rules, and financial constraints.

Inside many corporates, early AI initiatives were exploratory. Pilot projects generated insights and internal learning, but they did not always change how the business actually ran. More recently, the emphasis has shifted toward embedding AI directly into operating systems, linking it to decisions that affect productivity, cost, and risk across entire business units.

In this article, I want to showcase three examples of corporate AI in action. Each case shows how artificial intelligence is being applied inside established agribusinesses, not as a side project, but as part of how the business now operates.

Counting What Counts

In Brazilian cattle feedlots, a surprising amount of management still depends on manual observation. Animal counts, feed trough readings, and behavior assessments are often based on human judgment. Small inaccuracies may seem harmless, but over a 120-day confinement cycle they compound quickly into measurable cost.

Cargill’s Cattle View system starts with a straightforward premise: better information leads to better decisions. Drones capture high-resolution images of the feedlot. Those images are uploaded to the cloud and analyzed using AI-powered computer vision models that count animals, assess distribution patterns, and monitor activity around the troughs.

One of the key outputs is Cargill’s Animal Welfare Index, known as the IBE, scored from 1 to 10. The index incorporates variables such as the percentage of animals standing or lying down, proximity to the trough, and position during feed readings. The operational target is to keep the score above 9.

But the IBE is more than a welfare indicator. Scores above 9 have also been associated with around $8 per head in additional margin. Improvements in counting accuracy, tighter control of feed consumption variation, and better inventory management contribute directly to financial performance. In this context, AI becomes part of the daily management cycle, shaping both animal welfare and profitability at a scale that would not be feasible through manual observation alone.

Into the Weeds

Agronomic decisions are rarely simple. Weed pressure, disease risk, weather patterns, crop genetics, and input costs interact across a growing season. Many of these decisions have traditionally relied on field scouting, historical experience, and fixed application schedules.

BASF’s approach adds artificial intelligence directly into that decision process. Digital mapping tools generate field-level data on biomass and variability. AI models integrate agronomic logic, crop development stages, meteorological inputs, and target biology to guide decisions around herbicides, fungicides, and growth regulators.

In weed management, variable-rate control has shown up to 60 percent optimization in resource use while maintaining or improving productivity. Growth regulators applied at variable rates have demonstrated around 30 percent input optimization alongside measurable yield gains. These adjustments translate directly into cost per hectare and input efficiency.

Disease timing offers another example. Applying fungicides too early increases cost. Applying too late risks significant yield loss. BASF’s AI-driven decision models combine meteorology, crop genetics, pathogen biology, and historical field data to identify the preventive application window. In comparative testing, the system achieved 98.7 percent performance in supporting optimal preventive decisions relative to field specialists.

Small shifts in timing translate into meaningful changes in cost and yield over a full season.

Machines in Sync

If Cargill shows what AI can do in animal protein, and BASF in crop protection, CNH represents the third layer of the system: machinery. Equipment sits at the center of farm operations. When machines operate in isolation, small coordination issues show up in planting windows, spraying schedules, and harvest timing.

CNH’s connected farm model builds on years of telematics and machine monitoring. Equipment transmits operational data in real time, including field activity, input application, fuel use, and performance metrics. Artificial intelligence analyzes those streams to highlight bottlenecks, adjust routes, and support coordination between machines working in the same field.

The effect is visible in core operating numbers. Recent seasons showed productivity above regional and national benchmarks at CNH's model farm in Mato Grosso. Machine availability improved through better maintenance planning. Harvest efficiency increased, reducing the number of days required to complete operations. Fuel consumption declined, lowering cost per hectare and reducing emissions intensity.

For farm managers, that translates into fewer surprises during critical windows. Better visibility into how equipment is performing allows earlier adjustments, keeping operations closer to plan and costs under tighter control. Here, AI is applied to something fundamental: keeping machines, timing, and cost aligned during critical field operations.

Smart at Scale

Across these three examples, the pattern is clear. In feedlots, better counting improves both animal welfare and margin. In crop protection, more precise timing reduces input waste and protects yield. In machinery fleets, tighter coordination lowers fuel use and shortens operational windows.

In each case, the starting point is visibility. When managers can see more clearly what is happening, whether in a pen of cattle, across a field, or inside a fleet of machines, they make smaller, earlier adjustments. Over time, those adjustments affect cost per head, cost per hectare, and cost per hour of operation.

None of these changes depend on a single breakthrough. They come from using better data more consistently. The benefits show up in steadier performance, tighter cost control, and fewer operational surprises during critical windows.

Large agribusinesses tend to move carefully. When they decide to integrate a system into everyday operations, it usually reflects sustained results rather than experimentation. In animal protein, crops, and machinery, AI is increasingly part of that operating discipline.

Cargill, BASF and CNH will present these cases at The Yield Lab Latam AI in Agro event this week (26/2) in São Paulo, in partnership with the Brazil Rural Society (SRB). The event will be held in Portuguese. In-person places are already fully booked, but online access remains available. You can register here

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Reality Bites

In a recent article, I argued that artificial intelligence would first change agriculture by improving decisions, not by automating farms. That remains true. AI’s earliest impact is cognitive. It reduces uncertainty and will help farmers make better calls under pressure.

Shortly after publishing that piece, however, I listened to an episode of Moonshots featuring a wide-ranging discussion with Elon Musk on AGI timelines and humanoid robots. The vision was extremely bullish, particularly for industries where capital is cheap, assets turn over quickly, and labor is scarce.

Agriculture in Brazil operates under a different set of constraints. High interest rates, long-lived machinery, legacy investments, and still-available labor slow the pace of structural change. That does not delay AI’s impact. It changes where it shows up first.

On Brazilian farms, AI will not arrive as a sweeping technological overhaul. It will arrive where uncertainty is most expensive and mistakes are hardest to absorb. Capital decisions. Production risk. Market exposure. These are the pressure points where better information and better timing can deliver value, without requiring large new investments.

Finance Friction

Capital is where AI first starts to matter in Brazilian agriculture, because that’s where uncertainty is most expensive. Credit decisions still rely heavily on blunt filters such as past credit history, generic risk buckets, and hard collateral like land titles that are slow and costly to enforce. What is often missing is a precise view of production risk, even though future production is the real collateral behind most farm loans.

In grains, TerraMagna applies satellite imagery, historical crop data, and AI models to assess the quality of future production as collateral. Founded by ITA graduates with strong geospatial and AI backgrounds, the company focuses on measuring and predicting production conditions with far greater precision than traditional credit models. It also estimates expected harvest windows, making repayment timing more visible and collection more predictable.

In coffee, Culttivo reflects the different risk profile of a perennial crop with a biennial cycle and a 15- to 20-year investment horizon. By combining satellite imagery, climate data, and ground-level validation, the platform distinguishes between low- and high-risk producers in a sector where farmers rarely walk away and irrigation has reduced climate exposure.

In animal protein, Agroforte builds risk profiles using production data already held by dairy, pork, poultry, and beef processors. Much of this information has historically been scattered across spreadsheets or filing cabinets. By structuring it, Agroforte helps identify higher-margin producers and supports investments in productivity and animal wellbeing. In many cases, processors are willing to help guarantee loans as a way to strengthen long-term supply relationships.

Harnessing Nature

If capital is where uncertainty is priced, production is where it is felt. Brazilian agriculture operates in open systems shaped by biology, weather, and timing, where static models struggle to keep up. As volatility increases, production risk is no longer just about yield. It is about resilience.

At the biological level, Symbiomics applies AI to the development of next-generation biologicals. By combining sequencing technologies, machine learning, and genome-editing tools, the company designs high-performance biological inputs that improve productivity while reducing reliance on synthetic chemicals. Instead of slow trial-and-error discovery, AI helps narrow the search space and accelerate learning.

Other risks are more immediate. In sugarcane, Umgrauemeio focuses on wildfire prevention, one of the sector’s most destructive threats. Using satellite imagery, high-definition cameras on monitoring towers, and AI models trained to predict where fires are most likely to start, the platform helps operators detect outbreaks in seconds and position resources closer to high-risk zones, reducing response time and damage.

Water risk follows a different logic. Kilimo combines sensor data and satellite imagery to build precision irrigation models that help farmers increase productivity while reducing water and energy use. Beyond efficiency gains, the platform enables farmers to participate in water stewardship programs, where savings can be monitored, reported, and verified, turning risk management into measurable income.

Money Matters

If production risk determines whether a crop exists, market risk determines whether it is profitable. Brazilian farmers operate in markets shaped by price volatility, basis risk, logistics constraints, and foreign exchange exposure. These factors are often treated as external forces. In practice, they are information problems.

AI begins to matter here by improving price discovery and origination. Grão Direto applies data and AI to connect farmers and trading companies more efficiently, improving transparency and deal formation in grain markets. By analyzing historical transactions, market conditions, and buyer behavior, the platform reduces information asymmetry and friction in commercialization.

Logistics is where market inefficiencies often compound. In Brazil, distance and coordination failures turn transport into a hidden tax on farm profitability. goFlux uses AI to improve the efficiency of agricultural logistics by matching freight demand with available capacity more intelligently. By reducing empty miles, delays, and coordination costs, logistics becomes more predictable, helping compress basis spreads and improve net prices at the farm gate.

Market intelligence follows a different path. Seedz has built a network of more than 150,000 farmers who earn loyalty points by registering activities and sharing data. AI transforms this large, real-world dataset into a business intelligence platform for input suppliers and machinery companies. What was once anecdotal feedback becomes structured insight, helping companies measure market share, identify demand trends, and spot commercial opportunities earlier.

Trust, First

Taken together, these three areas point in the same direction. AI is beginning to matter in Brazilian agriculture by reducing friction in the decisions that carry the most risk. It improves how capital is allocated, how production risk is managed, and how crops move into the market. In each case, the value comes from clearer signals and better timing.

This also explains why much of the global AI narrative struggles to translate to agriculture. Visions built around cheap capital, fast asset turnover, and extreme labor scarcity do not map neatly onto farming systems shaped by long-lived equipment, high interest rates, and biological cycles. In Brazil, those constraints shape where AI takes hold first.

What emerges instead is a slower, more practical path built on trust. AI earns its place by working quietly in the background, using data that already exists and improving probabilities rather than promising certainty. Over time, these gains compound. Better credit decisions support better production choices, which in turn provide better market positioning.

If you are interested in this topic, The Yield Lab Latam, together with the Brazilian Rural Society, will be hosting an event later this month on February 26 in São Paulo to showcase real case studies of AI being used in Brazilian agriculture. The session will feature three startups and three corporates, will be held in Portuguese, and will also be streamed online. Registration is free, although in-person places are limited, and available at the following link.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Born to Produce

Brazil does not need to prove it can produce food at scale. That chapter is already written. Over recent years, the country has cemented its position as one of the world’s most efficient and competitive agricultural producers, culminating in Brazil overtaking the United States as the world’s largest beef exporter.

That milestone was not the result of a sudden technological leap. It reflected decades of accumulated progress in genetics, tropical agronomy, farm management, and logistics, layered onto natural advantages of land availability, climate diversity, and the ability to produce across multiple seasons.

For much of that journey, success was measured in output. Producing more, producing reliably, and doing so efficiently defined progress. Capital, talent, and innovation flowed toward strengthening production systems on the ground, improving yields, reducing losses, and expanding acreage with discipline.

That focus delivered results. Brazil emerged as a global agricultural leader, capable of competing across crops, geographies, and export markets. The production race was won. What is changing now is how that production strength is converted into value.

New Dawn

Brazilian agriculture is already far more digital than it was a decade ago. Digital tools were first adopted to solve practical farm problems: managing scale, coordinating logistics, improving agronomic decisions, and operating efficiently across large and dispersed operations. This transition happened gradually and quietly, driven by necessity rather than strategy.

That quiet shift is now visible in the data. Surveys from the past two years show that more than 80 percent of Brazilian farmers use digital channels daily for farm-related activities, while over 70 percent rely on digital tools during their purchasing decisions, combining physical and online workflows. Digital payments in agriculture have roughly doubled in recent years, reaching more than half of producers, and nearly one in two farms now uses at least one precision agriculture tool, with a growing share operating multiple systems in parallel.

What has changed is not the presence of technology, but its role. Tools that once supported day-to-day decisions in the field are increasingly being used to organize information, reduce administrative friction, and support interactions beyond the farm gate. Digital systems are moving from operational aids to the underlying infrastructure of the farm business.

This shift reflects a broader transition in priorities. As production gains have matured and margins tightened, attention has moved beyond yield toward how performance is recorded, compared, and carried forward. Digital adoption is no longer only about farming better. It is about positioning farms to capture more value from what they already produce.

Hard Proof

Modern farms in Brazil generate enormous volumes of information. Machinery, inputs, consultants, climate models, and daily field operations all produce data. Yield maps, application records, animal performance metrics, and management decisions are already part of routine farm life.

The challenge is not producing information, but organizing it. In many operations, data remains fragmented across spreadsheets, PDFs, emails, and disconnected platforms. Records are created for immediate use, then set aside, rather than maintained as part of a continuous operational history.

As a result, farms often perform better than they appear on paper. Performance is clear internally, but difficult to present externally. Banks, insurers, and buyers frequently see partial snapshots instead of a coherent picture of how risk is managed and outcomes are delivered over time.

This is where value is captured or lost. Not because farms underperform, but because performance cannot travel easily beyond the farm gate. Digital systems that consolidate records, reduce manual work, and maintain continuity turn output into evidence. They make results usable, comparable, and defensible in markets that increasingly demand proof.

Why Now

For years, this disconnect carried limited consequences. Today, it does not. The environment around agriculture has changed, and tolerance for opacity has narrowed. Capital is more selective. Insurance pricing is more granular. Buyers face tighter requirements around traceability and compliance. Sustainability commitments increasingly require verification rather than intention.

Volatility amplifies the effect. When margins tighten and uncertainty rises, access to credit, insurance terms, and market channels becomes more important than incremental yield gains. Farms that can document performance move faster and negotiate from a stronger position. Those that cannot face higher costs, delays, or exclusion.

At the same time, the barriers to digital adoption have fallen. Tools are easier to use, require less manual input, and increasingly work despite connectivity constraints. Artificial intelligence is accelerating this shift by automating data capture and turning existing information into usable insight with minimal friction.

Digital adoption has become costly to ignore. The focus has shifted away from experimenting with new tools toward reducing friction, as value increasingly flows through verified information rather than assumptions.

Sharp Edge

Brazil has already done the hardest work. It built a production system capable of competing at global scale under demanding conditions. The next advantage comes from turning that strength into proof that travels across financial, commercial, and regulatory channels.

As farms move from output to evidence, they gain access to cheaper capital, better risk pricing, and broader market opportunities. Digital systems do not change how food is produced. They change how performance is seen, priced, and rewarded.

For producers operating in export-driven markets, the question is no longer whether this shift applies, but how long it remains optional. Brazil is already moving up the value curve, and those competing alongside it will increasingly need to do so on the market’s terms.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Ground Up

When people imagine AI in agriculture, they tend to picture robots in fields and fully automated farms. Those systems exist, but they are capital-intensive, fragile, and slow to scale. The earliest and most consequential impact of AI will arrive elsewhere, through better decision-making long before machines become ubiquitous.

Farming has always been defined by decisions made under extreme uncertainty. What to plant, when to plant, how much to invest, how to manage risk, and when to sell. These choices sit at the intersection of biology, weather, markets, and increasingly regulation and consumer expectations. The underlying challenge remains the same, while the tools available to manage it have improved materially.

Artificial intelligence is entering agriculture incrementally, starting with how decisions are made at the farm level. It enters through systems designed to reduce uncertainty in environments where mistakes compound quickly and outcomes cannot be reversed. Its impact comes from improving how information is gathered, interpreted, and turned into decisions that reflect how farming actually works.

Ripe Time

Agriculture’s cautious relationship with technology has always been rational. The cost of being wrong is high, while the benefit of marginal improvement often becomes visible only at the end of a cycle. Many earlier waves of agricultural technology added insight while increasing complexity. Tools demanded attention, interpretation, and behavior change at moments when farmers had little time to spare. Cognitive load rose faster than confidence.

What makes this moment different is a shift in underlying conditions. Data density has crossed a threshold. Machinery, satellites, sensors, genetics, and supply chains now generate continuous streams of information. Compute has become inexpensive enough to process this data in near real time. Interfaces are receding into the background, replaced by systems that return recommendations instead of reports.

Market pressure reinforces this shift. Buyers want traceability and proof of production practices. Capital demands measurable risk and defensible returns. Regulators are moving from voluntary disclosure toward enforceable reporting. Sustainability has shifted from intention to verification. Verification at scale is a data problem, and data problems increasingly require automated decision systems.

Friction Free

There is a useful historical parallel. GPS did not transform agriculture by producing new agronomic insight. It transformed agriculture by removing friction. Once precise location became cheap and reliable, practices such as spot spraying and variable-rate application became practical at scale. GPS expanded what was feasible.

A similar pattern is now emerging with tools like virtual fencing in dairy and grazing systems. Instead of installing kilometers of physical fence, farmers can define boundaries digitally and adjust them in real time. The appeal is not sophistication for its own sake, but simplicity: fewer materials, less labor, faster changes, and lower upfront cost. When a system removes effort rather than adding it, adoption follows.

AI plays a similar enabling role today. By collapsing the cost of collecting, processing, and interpreting data, it shifts the constraint from information availability to decision quality. Voice-based interfaces lower barriers of literacy. Messaging-style interaction, closer to WhatsApp than traditional software, fits naturally into daily routines. When decision support becomes this accessible, it stops behaving like a product and starts behaving like infrastructure.

Taking Root

This shift is already visible across the agricultural value chain. Planting and input decisions increasingly incorporate probabilistic weather and yield models rather than historical averages alone. Credit and insurance rely on dynamic assessments of biological and climate risk, shortening the distance between risk evaluation and capital allocation. Logistics systems adjust routing and timing for perishables as conditions change. Breeding, formulation, and production decisions are simulated digitally before capital is committed in the field.

Agriculture has always combined tasks and purpose. The tasks include scouting, recording, checking, comparing, and updating plans as new information arrives. The purpose is steering a biological system through uncertainty, protecting yield, quality, and margins as conditions shift. AI will absorb more of the repetitive work and simplify data collection and decision-making. The result is not the removal of the farmer, but greater leverage for the judgment that matters most.

These changes rarely announce themselves as technology adoption. They appear as quieter improvements in how decisions are made, tested, and refined. As these practices spread, they begin to reshape expectations around timing, precision, and accountability.

Fog Lifts

Uncertainty remains the hidden tax of the global food system. It inflates the cost of capital, encourages over-application of inputs, distorts pricing, and discourages long-term investment. As uncertainty becomes measurable, behavior shifts. Risk premiums compress. Capital moves earlier. Decisions improve incrementally, then structurally. The effect compounds across farms, supply chains, and markets.

When that change becomes visible, public markets are likely to be among the first places it shows up. Agricultural companies will no longer be priced purely as mature, volatile, and capital-intensive businesses with limited scope for structural improvement. As decision-making improves and uncertainty declines, growth looks different, risk is reassessed, and valuations begin to adjust accordingly.

Emerging markets make this dynamic especially visible. These systems operate under tighter capital constraints, higher volatility, and fewer margins for error. They cannot wait for perfect data or ideal conditions. Agriculture in these environments becomes a demanding proving ground for AI, shaped by biology, infrastructure limits, and real-world pressure. If decision systems function here, they function anywhere.

AI will not feed the world by producing food directly. It will feed the world by changing how decisions are made under uncertainty. That shift may prove more consequential than any single technological breakthrough.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

False Start

The first wave of AgTech struggled because it was built around individual tools and solutions rather than how farming actually works. Farmers and founders couldn't see the forest from the trees. AgTech was too narrow, too agronomic, and too clunky, solving problems in isolation and generating siloed data that added effort instead of real value or transparency.

I’ve often described Brazil’s agriculture as a black box, but even that comparison is generous. At least a black box records everything that happens and can be examined later if needed. Brazilian agriculture, for the most part, has never been recorded in any consistent or visible way. Data exists, but it is fragmented, incomplete, and rarely connected across decisions, seasons, or systems.

Many of the limitations in Brazil are structural. Connectivity is weak. Farms are complex and non-standard. Decision-making is spread across people, tools, and moments in the season. While these constraints have traditionally been difficult to overcome, recent advances in AI now make it possible to tackle multiple challenges at once instead of addressing them individually.

AI doesn’t just add intelligence on top of farming. It can quietly remove friction from underneath it. In a country like Brazil, where connectivity gaps, biological complexity, and trust pressures are part of daily life, that shift may finally unlock the kind of AgTech adoption the sector, and VC investors, have been waiting for.

Clean Slate

Brazil enters this next phase of AgTech with an unusual advantage. According to McKinsey, fewer than one in five Brazilian farmers use any form of digital farm management software, and usage tends to be fragmented and shallow rather than fully integrated. That gap is often described as a weakness, but it also means there are fewer legacy systems standing in the way of change.

Where these systems do exist, they are often treated as a chore rather than a tool. Data gets entered because someone else needs it, usually for certification or compliance, not because it helps farmers make better decisions. I recently spoke with a coffee farmer who had two people dedicated solely to feeding data into a homemade CRM. The system existed to satisfy certification audits, not to improve how the farm actually operated.

This matters more than it might seem. Legacy software creates switching costs, resistance, and long sales cycles. It also trains farmers to associate technology with extra work. In Brazil, that baggage is lighter. Many farms are effectively starting from zero, which creates an opening for AI-first tools that replace manual effort instead of adding another layer.

From a VC perspective, this speaks directly to the Why Now question. Market pressure around traceability, sustainability, and transparency has been building for years, but the ground truth was always hard to capture. AI-driven tools are now making daily work easier, with less training and fewer reasons to resist. When transparency becomes a byproduct of how the farm already runs, rather than an extra task, adoption starts to move much faster.

Broken Signals

Connectivity remains one of the biggest mismatches between how AgTech is built and how farming actually works in Brazil. Large areas still have weak or unreliable signal, and even farms with decent coverage can lose it at the wrong moment. Yet most software assumes constant connectivity and a user sitting still in front of a screen.

That assumption breaks quickly in the field. Farmers make decisions while walking crops, driving machinery, or checking animals. They do not want to stop what they are doing to type notes or fill out forms. When a tool depends on signal, logins, or structured inputs at the wrong time, it creates friction and quickly falls out of use.

This is where AI starts to reduce friction in a very practical way. Farmers already send voice notes and photos every day, often through WhatsApp. There is nothing new to learn. With Voice AI tools, the difference is that these messages can be captured even without signal, stored, and processed later. Data collection happens close to the moment of observation, instead of being reconstructed from memory later on.

That shift matters more than it sounds. Information is more accurate because it is captured in context. Farmers spend less time reporting and more time farming. For AgTech companies, the data is cleaner and more consistent. And for investors, this points to adoption that scales because the technology fits existing habits rather than asking farmers to change how they work.

Fast Biology

Farming in Brazil moves at a different pace. Crops grow quickly, pests and diseases never fully disappear, and there is no long off-season to pause and reset. Decisions made today can show consequences within days, sometimes hours. Waiting for perfect information is often not an option.

This is one reason traditional AgTech has struggled. Many tools were designed for slower systems, where problems unfold over weeks or months. Data is collected, uploaded, processed, and reviewed later, often too late to influence what actually happens in the field. By the time insights arrive, the window for action has already passed.

AI fits better in fast biological systems because it can work across many signals at once. Weather, crop stage, disease pressure, field history, and farmer observations can be considered together rather than split across separate tools. When photos and voice notes are captured as things happen, the system starts to reflect reality more closely.

Brazil’s year-round farming also creates constant feedback. Similar situations repeat across regions and seasons, giving AI more chances to learn and adjust. What looks complex on the surface is actually a rich environment for systems that improve through use. In this context, speed and variability stop being just challenges and start becoming an advantage.

Why Now

For a long time, AgTech struggled because adoption depended on persuasion. Farmers had to be convinced to change how they worked, learn new systems, and take on more reporting, often with unclear upside. Even when the technology made sense, the effort required to use it slowed adoption.

That balance is now shifting. AI-driven tools are lowering the cost, and ease, of usage. Voice, photos, and simple observations fit naturally into daily routines, which means less training, less resistance, and far less friction. Technology will no longer feel like a separate layer on top of farming. It will disappear into the background.

At the same time, pressure from markets and consumers is no longer abstract. Transparency, traceability, and sustainability are increasingly tied to access, pricing, credit, and insurance. For many farmers, providing proof is becoming part of staying competitive, not just meeting regulations.

This is a powerful combination that allows for change. Ease of use removes resistance, while market pressure removes choice. Together, they create the conditions for adoption to scale quickly rather than slowly. When doing nothing becomes harder than using the tool, behaviors change.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Behind the Curtain

They say travel broadens the mind. That proved very true for the ARISE* trilateral delegation as they touched down in São Paulo last week, expecting to see tropical agriculture but instead finding something far more layered. Over the course of the week, the group — made up of participants from the UK, US, and countries across Latin America — discovered a sophisticated agri and climate innovation ecosystem operating largely out of sight of global limelight.

The same sentiment was felt earlier this month at COP30, where foreign visitors were stunned by Embrapa’s Agrizone — a walk-through demonstration of over fifty net-zero crops that revealed a Brazil they did not know existed. It was a reminder that much of the country’s innovation strength is visible only when you step inside it.

That became the theme of the ARISE week. The delegation spent their days moving through São Paulo’s own agri zones — scientific, industrial, and environmental — seeing firsthand what rarely appears on international radars. Three visits in particular reshaped their understanding. Embrapa showed a country reinventing agricultural research for a digital and climate-conscious era. CTC demonstrated an energy transition that has been running at national scale for decades. And CNPEM revealed a frontier-science capability few expected to find.

Embrapa’s Evolution

Any exploration of Brazil’s innovation system begins with Embrapa. Created to solve the challenge of farming in tropical conditions, Embrapa played a central role in transforming the Cerrado — once considered unsuitable for agriculture — into one of the world’s most productive grain regions. Its breakthroughs in tropical seed varieties, soil correction, and systems agronomy underpin much of modern Brazilian agriculture. Yet Embrapa today is far more than the institution that unlocked the Cerrado. It is a distributed network of more than forty specialised research units, each advancing a different frontier of agri and climate innovation.

The ARISE delegation engaged with two of these units in Campinas. Embrapa Digital Agriculture, situated within the prestigious Unicamp campus, is building the digital foundations of tomorrow’s farming — data modelling, sensor networks, geospatial analysis, and climate intelligence that help producers navigate complexity at scale. A short distance away, Embrapa Environment focuses on the ecological dimension: biodiversity, water quality, bioinputs, soil health, and land-use transitions. Together, they reflect how Embrapa blends technological and environmental insight into a unified vision of sustainable production.

AgNest, Embrapa’s farm lab on the outskirts of Campinas, showed how this vision is tested in the real world. It is a fully operational farm designed for technology trials under commercial conditions, where startups, researchers, and industry partners work side by side to evaluate what truly delivers impact in the field.

Across these touchpoints, the group gained a clearer sense of Embrapa’s breadth — and of how actively it is shaping Brazil’s next chapter of agricultural and climate innovation.

Bioenergy Engine

Few countries have a renewable-fuel legacy as deep as Brazil’s. Its sugarcane ethanol program began in the 1970s, giving it nearly fifty years of accumulated expertise, with CTC serving as one of the key centres moving this innovation engine forward.

What the ARISE delegation encountered at CTC was an industry built on steady, cumulative innovation. High-performance cane varieties, precision breeding, efficient milling systems, bagasse-to-energy loops, and emerging pathways for second-generation ethanol all reflect decades of coordinated scientific and industrial effort. The result is a renewable transport ecosystem operating at national scale, one that has quietly shaped Brazil’s emissions profile for generations.

CTC’s influence extends well beyond Brazil’s borders. Countries across Africa, Asia, and Latin America license Brazilian genetics, adopt its processing technologies, and study its policy frameworks when designing their own bioenergy programs. CTC itself has also maintained a research presence at the Donald Danforth Plant Science Center in St. Louis for several years, deepening the international exchange around sugarcane science and bioenergy innovation.

For the ARISE group, the visit underscored how deeply agriculture, industry, and climate are intertwined in Brazil. Sugarcane is simultaneously a crop, an energy source, a climate asset, and a platform for biotechnological innovation — a reminder that the energy transition here is not aspirational, but already embedded in daily life.

Frontier Science

CNPEM, the Brazilian Center for Research in Energy and Materials, was the visit that most shifted the group’s sense of Brazil’s scientific capacity. Many arrived expecting an advanced agricultural laboratory. Instead they found a national research campus with capabilities on par with leading centres in the US and Europe, anchored by Sirius — one of the most sophisticated synchrotron light sources in the world.

Sirius allows researchers to observe biological and material structures at near-atomic resolution, opening the door to breakthroughs in plant science, pathogen resistance, soil–root interactions, biomaterials, and engineered molecules. Surrounding it are four national laboratories covering biosciences, nanotechnology, renewable materials, and biorenewables. Together, they form an integrated platform where questions that span agriculture, climate, and energy can be investigated at a level of detail few countries can match.

For the ARISE delegation, the scale and sophistication of CNPEM came as a surprise. The facilities were not just impressive; they were deeply relevant to the challenges facing global food and climate systems. The experience reframed Brazil not only as a producer or a testbed, but as a contributor to frontier science with the tools to shape how the next generation of technologies is developed.

Next Steps

The first ARISE Knowledge Exchange Week was built around discovery, and Brazil delivered it at every turn. Embrapa revealed the depth of scientific and environmental thinking behind the country’s agricultural transformation. CTC showed how a crop became the foundation of a national renewable-fuel system. And CNPEM demonstrated that Brazil is not only producing at scale, but also generating frontier science with global relevance. Each visit offered a different lens, yet together they lifted the veil on a country whose innovation capacity is far greater than its public narrative suggests.

This was only the opening chapter. In February 2026, the ARISE delegation will travel to St. Louis to explore another critical part of the global innovation ecosystem, from the Danforth Center and Taylor Geospatial Institute to the region’s industrial biotech networks. A month later, the final stage in Cambridge will bring the UK’s translational infrastructure and engineering biology expertise into the conversation. São Paulo helped the group see what already exists. St. Louis and Cambridge will help shape what might come next.

*ARISE (Agri-Tech for Resilience, Innovation & Sustainable Ecosystems) is a UK Government–funded Tactical Fund project led by Agri-TechE, in partnership with AgriTIERRA, Earthbase, Embrapa, and The Yield Lab Institute. The São Paulo Knowledge Exchange Week brought together participants from NIAB, Newcastle University, Harper Adams University, the Donald Danforth Plant Science Center, the Taylor Geospatial Institute, Washington University in St. Louis, CREA, University Austral, AgCenter Colombia, Tec de Monterrey, AgTech Chile, and The Yield Lab Latam.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Rural Reality

Many AgTech founders eventually hit the same wall. Their idea seems interesting, early pilots look promising, and investor decks indicate a huge market opportunity. Yet the commercial curve never takes off. Meetings drag on for months, pilots stretch over multiple seasons, and widespread adoption remains elusive.

Last week I wrote about how market timing determines the success of startups in general (Mind the Gap). This week, I want to go deeper into the specifics of agriculture, where timing is even more complex and stakes are much higher.

Farmers are not slow to adopt technology because they are risk averse. They take risks every day and are the original entrepreneurs. What slows them down is the system around them — weather, price volatility, labor shortages, and constant uncertainty. Every decision flows through these filters. Startups often arrive assuming technology will be judged on performance or innovation. In practice, it is judged on risk, familiarity, cash flow, and trust.

Three forces in particular shape how farmers decide what to adopt and when: the economics of uncertainty, the familiarity barrier, and the first-experience effect. Together they form the roots of resistance that most AgTech founders underestimate, and that ultimately determine whether a company grows or stalls.

Margin Math

Farming is one of the few industries where a bad week of weather, a sudden pest surge, or a market drop can erase a year’s work. Most farmers operate on thin margins. Their budget resets every season, their revenue is volatile, and their cost structure is increasingly unpredictable. Under these conditions, every investment is filtered through one question: will this pay back in the current harvest season?

AgTech founders often assume their tools should be viewed as multi-year investments. Farmers rarely think this way. They make decisions through the lens of inputs: fertilizer, pesticides, seeds, fuel. These are annual choices. If a technology does not fit this seasonal logic, or if the payback stretches over several years, the adoption window narrows quickly. Even the most promising tool starts to feel like a luxury.

For farmers with predictable margins, such as irrigated fruit, high-value crops, or dairy under stable contracts, multi-year investments make sense. For the many who farm grains or livestock under tightening cycles, the math is unforgiving. If the numbers do not close this year, the decision is postponed. And once postponed, it may not come back for a long time.

This is not a question of ambition or conservatism. It is simply the financial reality of running a farm within a system defined by uncertainty. Startups that overlook this disconnect often end up with strong products but slow commercial traction because the economic environment shaping farmer decisions does not match the assumptions behind most venture-backed growth plans.

Familiarity Factor

Even when farmers can afford to adopt new tools, another barrier often slows the process: unfamiliarity. Most people evaluate new technology by comparing it to something they already know. Farmers follow the same logic. They want to understand who else is using it, what results were achieved, how it fits into a routine, and what happens when something goes wrong. Without that context, it is almost impossible to estimate the real return on investment.

This is why AgTech adoption depends heavily on peer validation. A single reference farm carries more weight than a hundred slides. Farmers want to see a machine running under real conditions, not simulated trials or ideal scenarios. They want to talk to someone who has used the tool through a full season, faced setbacks, and knows whether support teams actually respond when issues appear.

Startups often underestimate how much effort it takes to create these reference points. Pilots move slowly because trust is earned across multiple seasons, not in a single demonstration. Agronomists must see performance hold up under different weather patterns. Co-ops want consistent data before recommending anything to their members. Each step in this chain takes time, and skipping any part of it usually backfires.

When familiarity is missing, even strong technology loses momentum. It does not move forward or backward. It simply stays on the edge of consideration because the farmer lacks the information needed to form a clear mental model of how it would work on their farm. Without that clarity, adoption rarely begins.

Starting Signal

If the economics of uncertainty and the familiarity barrier slow adoption, the first experience becomes the signal that shapes everything that follows. Farmers remember their early encounters with new tools. A poor installation, unreliable performance, or weak support quickly becomes a story retold at co-ops and family gatherings for years.

This pattern has roots in AgTech’s early years. Many tools reached the farm before they were ready for real conditions. Startups underestimated the complexity of day-to-day operations, stretched their support teams thin, and promised more than the product could consistently deliver. Solutions that looked impressive in demos struggled to perform in the dust, heat, and unpredictable rhythms of actual farm work.

The reverse can also happen. My brother’s adoption of robotic milking is a good example. For a long time, the benefits were clear, but the concerns carried more weight. The system was expensive, and there were questions about how the cows would adapt. It took a new and very practical challenge, the question of succession, to move the decision forward. Once the system was installed, it worked better than expected. That first exposure changed his outlook entirely, and now he looks for ways to automate every part of the farm.

On most farms, adoption depends on a handful of defining moments rather than a steady transition. One successful pilot encourages further investment, while one bad experience can slow decisions for years. Startups that understand this dynamic focus their early deployments on farmers who enjoy testing and troubleshooting. These early adopters know that new tools may have rough edges and treat those moments as part of the process.

Final Footnotes

Slow technology adoption on the farm is not necessarily due to inertia or fear of change. It often reflects the realities of a system defined by uncertainty. Most farmers operate on thin margins, so every decision is filtered through the maths of survival. When a tool is unfamiliar, they lack the reference points needed to evaluate it properly. And first experiences carry outsized weight, shaping expectations for years.

Other frictions add to this. Even simple tools require new routines, training, and adjustments to existing workflows. Compatibility issues with legacy machinery and fragmented systems add another layer of caution. This is one reason why developing markets sometimes adopt technology faster. Without the weight of older infrastructure, they can leapfrog straight to newer solutions that are harder to introduce in more established systems.

For founders, these realities are not obstacles to dismiss. They define the environment in which AgTech must grow. Success depends on understanding the financial rhythms of the farm, sequencing early pilots carefully, and earning trust through consistent performance. The companies that scale will be those that build around the way farmers really work and make decisions. AgTech moves in seasons, not sprints, and the startups that embrace this pace give themselves the best chance to thrive.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Right Time

Timing is the least glamorous part of a startup story, yet it’s often the most decisive factor — and the most overlooked. Teams work obsessively on technology. Investors make big bets on talent. But the market has its own clock, and it rarely tells you what time it is.

Bill Gross, the founder of Idealab, discovered this the hard way after launching dozens of startups across different industries. Some thrived. Many didn’t. Eventually he ran a study comparing factors across a few hundred companies — team, idea, funding, business model, and timing. Timing dominated them all. A good idea at the wrong moment failed just like a bad one.

That doesn’t mean execution doesn’t matter. It just means markets open and close on their own schedule. Some founders hit the window perfectly. Others build brilliant products in years the market simply isn’t ready.

Technologies that look obvious in hindsight often spent years gathering dust because the ecosystem wasn’t there yet. Others explode almost overnight because a pressure point, a trigger, or a shift suddenly opens the door.

Every founder wants to build the future. The real test is whether the future shows up while they’re still in the driver’s seat.

Early Bird

Some companies don’t miss the future — they meet it too early. Kodak is the textbook case. Long before we carried cameras in our pockets, a young engineer inside the company built the first digital camera prototype in 1975. The technology worked. It was clunky, slow, and expensive, but it worked. The problem was that the world had no use for it yet. Retail photo labs, printers, storage, sensors — none of it existed at consumer scale. Kodak had the right idea, just delivered into a market that wasn’t remotely ready.

When a company gets burned by being early, it often becomes blind to the moment when the world finally catches up. Kodak fell into that trap. As digital imaging became viable decades later, they hesitated. They had already been bitten and didn’t want to bet the company again.

QR codes followed a similar arc, though with a luckier ending. Created in 1994 to track parts in Japanese factories, they spent years bouncing around marketing teams, always clever but never critical. People saw them, but no one needed them. Then COVID forced restaurants to eliminate physical menus, and a quiet, underused square suddenly found its purpose. The technology didn’t improve; the context did. Timing finally clicked.

Blind Spot

Even when founders build something genuinely useful, they often misread who actually cares enough to pay for it. It’s one of the most common and costly blind spots in early-stage companies: assuming the user is the customer, or assuming the person with the problem is the same person with the budget.

Our Argentine portfolio company, Kilimo, learned this lesson early on. Their platform helped farmers irrigate more efficiently, save water, and improve decision-making. Farmers liked the insights. Agronomists appreciated the data. But the urgency wasn’t there. Water savings were valuable in principle, yet not painful enough in practice to trigger rapid adoption or meaningful spending.

Meanwhile, a very different group was feeling real pressure: corporate sustainability teams. Auditors, investors, and global buyers were demanding verified water data, risk assessments, and credible reduction claims. These companies weren’t looking for agronomic improvement; they were looking to satisfy compliance requirements and avoid reputational headaches. Kilimo didn’t change its technology — it changed its angle. The same water insights became the foundation for a verified water-offset program, sold to buyers upstream rather than farmers downstream.

Market Watch

Sometimes markets don’t evolve gradually — they snap open when a single enabling force arrives. That force can be technological, like GPS, which quietly transformed what was possible long before anyone realized it. Before accurate positioning lived inside every phone, ideas like real-time navigation, ride-hailing, food delivery, or on-demand logistics were simply unbuildable. The concepts weren’t new; the infrastructure wasn’t there. Once GPS became universal and reliable, an entire generation of companies suddenly had a runway.

E-signatures existed long before 2000, but adoption stayed stuck because they weren’t legally recognized. The moment the U.S. ESIGN Act and the EU’s eIDAS rules came into force, digital signing went from fringe to mainstream. This single regulatory shift unlocked the rise of DocuSign, Adobe Sign, and a wave of workflow platforms that had been waiting patiently in the wings. A market that had been stagnant for a decade transformed almost overnight because the law finally caught up with the technology.

In both cases, timing wasn’t about founder brilliance. It was about sensing when the environment had finally tipped in favor of new behavior. One new technology or one new regulation can redraw the map overnight, opening an opportunity that didn’t exist the day before. Founders who spot these changes early are the ones who get to build while everyone else is still adjusting.

Mind the Chasm

Getting market timing right is hard because most of the forces that shape it sit outside our control. You can build something smart and still arrive years before anyone cares. When that happens, you either wait or shift your focus until the world catches up — but in both cases you need to watch the signals carefully: where pressure is rising, where incentives are shifting, and where the system is starting to open.

If product–market fit isn’t clicking, widen the search. The most obvious user isn’t always the one who feels the pain or holds the budget. Geoffrey Moore talked about the chasm between early adopters and the mainstream; the same logic applies to customers. Sometimes the real market sits a few steps away from where you first aimed.

And be cautious when chasing problems in markets you don’t fully understand. Hidden incentives and entrenched habits can quietly kill a solution long before the technology gets a chance. Innovation only sticks when it fits the unwritten rules of the industry you’re trying to change.

The companies that win aren’t just good at building. They’re good at sensing when a market is shifting — a new technology becomes accessible, a rule changes, or a behavior suddenly becomes normal. These moments don’t announce themselves, but they open new markets for the teams paying attention.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Change Ahead

For generations, farming has been a sensory craft. A quick squeeze of the soil or the sound of a tractor running differently told farmers much of what they needed to know. Instinct became a competitive advantage. Experience became a sixth sense. Farming rewarded those who could read subtle cues and make good decisions in the face of uncertainty.

But the job is getting harder. Farms are larger, climates swing more sharply, and every decision carries more weight — with less margin for error. What once relied on observation and timing now demands a level of consistency and precision that human senses were never built to deliver. Even the best operators can only track so much across a busy season.

That’s where technology is stepping in — not to replace intuition, but to extend it. Tools that measure, record, and map the farm give farmers a wider field of view, revealing dynamics that were once easy to miss. The craft of farming remains, but the way we perceive the farm is expanding.

Data Storm

Farming has always looked simple from the outside — a field is a field, a crop is a crop. But beneath that apparent uniformity lies a dynamic system shaped by soil chemistry, canopy temperature, root development, microclimates, nutrient flow, and countless biological interactions. Most of these dynamics unfold quietly and invisibly, far beyond what human senses are able to pick up in real time.

Technology begins to widen the frame. Soil probes track moisture shifts through the day. Satellites catch early signs of stress long before a leaf changes colour. Weather stations capture hyperlocal patterns that never show up in national forecasts. Even tractors, sprayers, and harvesters now stream data as they work, turning routine operations into detailed maps of how the field behaves.

The result is an unprecedented flood of information. Chemical signals, temperature changes, soil movements, atmospheric shifts — all of it captured continuously and at a scale no person could ever monitor alone. A modern farm generates more data in a single season than previous generations gathered in a lifetime, and the gap between what’s happening and what we can perceive grows wider each year.

Connecting Dots

Capturing data is only half the job — making sense of it is where AI steps in. It pulls these signals together and shows how small shifts in moisture, temperature, soil chemistry, or canopy behaviour link up to influence outcomes. When moisture readings sit beside canopy temperature, or when nutrient maps align with satellite imagery, new patterns appear — the kind no farmer could spot with the naked eye.

These systems learn from the farm itself. They notice subtle trends farmers don’t have time to track, integrating historical performance, weather forecasts, sensor data, and real-time field signals. A system might pick up, for example, that a slight rise in canopy temperature on a particular slope consistently follows a dip in soil moisture — a pattern invisible to even the most attentive scout. Over time, the machine begins to anticipate issues before they appear, offering recommendations that feel less like analysis and more like instinct — only with the accuracy of high-resolution sensing behind them.

The result is a shift from reactive decisions to proactive ones. Farmers stop responding to what they can see and start preparing for what they can’t. Technology doesn’t override farmer wisdom; it enhances it, giving producers a kind of augmented awareness that blends experience with machine-level sensing power.

Second Sight

The next shift in perception won’t come from dashboards alone. As tools evolve, the goal is to bring insight closer to where decisions happen — in the cab, on the quad bike, or walking a field. Augmented displays, guided overlays, and wearable interfaces can turn complex data into something almost physical, letting farmers glance at a landscape and see moisture zones, nutrient hotspots, or emerging pest pressure layered over the areas they already know.

Much of this happens automatically, without the farmer needing to stop and check a screen. Connected equipment, autonomous scouts, and always-on monitoring networks register changes that would otherwise slip by unnoticed. Instead of searching for problems, farmers receive targeted nudges: a subtle anomaly in one corner of a field, a shift in vigour along a slope, a zone drying out faster than expected.

As these tools mature, perception stops being limited by the human body. The farm becomes easier to “read,” not because a person sees more, but because the systems around them surface the right clues at the right moment. It feels less like using technology and more like gaining a new sense that fits naturally into the rhythm of work.

Wider Horizons

Extended perception doesn’t replace what farmers already know — it builds on it. The sharp eye, the gut feeling, the experience shaped by years on the land all remain essential. What’s changing is the range of what can now be sensed. The invisible becomes visible, the complex becomes legible, and the field itself becomes easier to understand.

And as perception expands, the profession itself is shifting. Farming was once a vocation more than a business — a way of life passed from one generation to the next, full of uncertainty and hard work but grounded in a deep connection to nature. Much of its value was non-monetary: the feel of soil between your fingers, the smell of fresh cut grass, the satisfaction of coaxing life from the land. Technology doesn’t erase that heritage, but it is nudging farming closer to an industrial model, one where efficiency and predictability carry more weight. Fields remain open-air factories — exposed to weather, biology, and risk — yet they are becoming more structured, more measurable, and more manageable.

This shift brings trade-offs. Farmers may spend less time physically immersed in the land, but they gain the ability to run their operations with more clarity, less waste, and far fewer blind spots. The volatility that once defined the business softens as better insight reduces risk and allows farmers to capture more upside across the supply chain. It’s a different kind of connection to the land — less tactile, more informed — but one that strengthens resilience and helps secure the future for the next generation.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Ground Zero

Every few years, the world resets its climate clock. Countries submit new plans under the Paris Agreement, known as NDCs, or Nationally Determined Contributions, outlining how they will cut emissions and adapt to a warming planet.

This year feels different. COP30, which will take place in Belém from November 10 to 21, 2025, will unfold on Brazilian ground, farmland that sits at the center of both global food production and the climate challenge. Roughly half of Brazil’s emissions come from land use and agriculture. If that sector does not shift, no number of electric cars or solar farms will close the gap.

As global leaders gather in the Amazon’s gateway city, the conversation around agriculture has changed. What used to be treated as a side issue about yields and exports is now a central pillar of climate strategy. Brazil’s farmers, once cast as the villains in deforestation debates, are being recast as potential heroes in carbon capture and land restoration. It’s a big shift, one that could reshape how Brazil is seen abroad and how farming works at home.

Tech Roots

Brazil’s updated NDC sets out its boldest climate plan yet: ending illegal deforestation in the Amazon by 2030, restoring 15 million hectares of degraded pasture, and expanding integrated crop, livestock, and forest systems by another 5 million. It also aims to capture more carbon in soils, turning farmland into part of the climate solution.

It’s the most ambitious agricultural commitment the country has ever made. But making promises is easier than keeping them. The same pastures that built Brazil’s commodity empire now account for a quarter of its greenhouse gas emissions. Fixing that will not come from government decrees alone. It will require coordination across ministries, banks, startups, and farmers, aligning credit, incentives, and technology so that sustainability pays.

That is where Brazil’s AgTech ecosystem comes in. Startups developing soil carbon sensors, satellite analytics, and digital farm management tools are laying the groundwork for a new level of transparency. The country’s success will depend on MRV: measurement, reporting, and verification, which in turn depends on digitization at scale. Without reliable data from farms, it’s impossible to track carbon gains, certify practices, or attract the capital these transitions need.

AgTech makes that possible. It connects fields to finance, translating what happens in the soil into data that banks, buyers, and regulators can trust. These tools are quickly becoming the backbone of Brazil’s climate strategy, not add-ons, but core to how the system works.

If Brazil can measure and verify carbon and land restoration at scale, it will not just hit its NDC targets, it could monetize them. Verified data becomes a new kind of asset, flowing through carbon markets, credit programs, and trade deals. That is how a climate promise starts to turn into a business case.

Hot Seat

Hosting a COP brings prestige, but also pressure. When the world meets in Belém, Brazil will need to show more than ambition. It will need results: hectares restored, methane curbed, credit aligned with climate goals. The government’s new climate-agriculture plan ties these pieces together, bridging what used to be separate silos of farm policy, environmental protection, and rural finance.

That integration is long overdue. For decades, agricultural growth and conservation were seen as conflicting goals. Subsidies rewarded expansion, not regeneration. Compliance rules punished farmers but rarely helped them change. The new approach flips that logic: reward performance, measure outcomes, and make climate-smart farming the norm, not the exception.

If it works, Brazil could become the world’s case study for tropical decarbonization, proof that a global breadbasket can also be a carbon sink. The timing helps. Global buyers are tightening import rules on deforestation and emissions, and food companies are scrambling to verify their supply chains. What once felt like compliance is now a competitive edge. The idea of sustainability as a cost is fading fast, replaced by a simple reality: in global trade, it’s fast becoming the price of entry.

Money Talks

Every NDC looks tidy on paper. The real question is who pays. Meeting Brazil’s land-use goals will take tens of billions in investment, but the financing architecture is starting to take shape. Fiagro funds, carbon markets, and sustainability-linked loans are turning policy ambition into private-sector opportunity.

Think of soil carbon as a new form of collateral. The more carbon a farmer stores, the better the credit terms. Satellites confirm pasture recovery, algorithms verify compliance, and carbon credits become tradable assets. It’s not theory, pilots are already underway.

The Cerrado region offers a glimpse of what’s possible. Once dismissed as degraded pastureland, it’s now home to innovative projects that mix crops, cattle, and trees, regenerating soil, cutting emissions, and improving yields. Farmers are earning premiums for practices that used to cost them money.

But the green transition has to include everyone. Large farms have the capital and technology to move first. Smaller producers need access to credit and training to follow. If climate finance stops at the top of the pyramid, Brazil’s climate leadership loses its moral base. A real transition means innovation and inclusion growing side by side.

A New Leaf

For decades, Brazil’s agriculture story was about expansion. Now it’s about renewal. If the goals in its NDC become measurable, financed, and farmer-led, the country could move from climate risk to climate leadership.

When the delegates fly home from Belém, the real work will begin, not in meeting halls, but in the red soils of Mato Grosso and the sandy pastures of Bahia. Progress will depend on what happens in those fields, how fast farmers can adopt new tools, access fair credit, and prove results that count.

The next chapter of Brazil’s climate story will not be written in reports or speeches. It will be written in data, in soil, and in the decisions farmers make each season. Growth and regeneration can share the same ground. The world will be watching to see if Brazil can show how.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

Ground Differences

For most farmers, technology isn’t about comfort — it’s about continuity. They adopt it to solve the problems that threaten their ability to keep farming. But the problems farmers face — and what they need technology for — change from place to place.

In developing countries, the toughest challenges often lie beyond the farm gate: access to credit, infrastructure, logistics, markets, and reliable data. In developed markets, the pressures sit inside it — labour shortages, ageing farmers, limited land, unpredictable weather, and questions of succession.

That difference explains why farm innovation takes such different forms from one country to another. In Brazil, technology is focused on connection — building the rails that link farmers to finance, buyers, and information. In Ireland, it’s about continuity — keeping family farms viable when hands are few and hours are long. Both are stories of necessity, just told in different languages.

Generation Change

I was back home in Ireland over the past two weeks, visiting family in County Monaghan. The landscape there is made of low rolling hills and heavy grey soil — not great for crops, but perfect for grass, which makes it ideal for dairy and beef. It’s a region shaped by livestock and weather, where herds and hedgerows mark time better than any calendar.

My grandfather started farming here in the 1930s with horses and a few cows, mostly for subsistence. My father took over in the 1960s, expanded the herd, bought tractors, and installed a herringbone milking parlour. Each generation added a layer of technology and a bit more business thinking.

Now my brother runs the farm. He’s kept the same fields, the same rhythm, but the tools have changed. The barns are wired with sensors, and the hum of the milking robot has replaced the clank of machines. What began as survival has become strategy — using technology not to grow bigger, but to stay viable.

New Routine

Walking through the yard, the changes are everywhere. The old parlour stands beside a small room where the robot milks cows quietly around the clock. Calves drink from automatic feeders that scan their tags and ration milk. The floors are scraped clean by another slow-moving robot that never complains.

The cows set their own pace now, wandering in for milking when they choose. The farm feels calmer, cleaner, and more efficient. But the work hasn’t disappeared — it’s just different. My brother spends more time checking dashboards than hoses, more time on maintenance than muscle. When something breaks, it’s usually a sensor, not a chain.

In Brazil, where labor is still more available and sunlight lasts all year, I often see the opposite challenge — too much land, not enough data. Farmers here are rich in natural resources but often constrained by credit, logistics, and information flow. Where Ireland’s bottleneck is labor, Brazil’s is connection.

On my brother`s farm contractors now handle the heavy work — silage, slurry, reseeding — which keeps the operation lean. Labour is scarce and expensive, and like many small Irish farms, succession is uncertain. My brother’s only son helps on weekends but isn’t sure he wants to farm full-time. Robots fill that gap. Once you see the time they save each day, there’s no going back.

Counting Cows

While home, I also visited Ireland’s agricultural research centre, Teagasc, in County Cork, where I met with researchers and advisors including Donagh Berry, Joe Patton, and Michael O’Donovan. Their work gives a glimpse into what’s happening beyond our own yard — and why even a highly digital dairy sector like Ireland’s still struggles with something as basic as accurate data collection.

Teagasc has some of the best pasture and feed models in the world. They can forecast growth rates and nutrient needs with impressive accuracy — at least on paper. The problem, the researchers explained, isn’t the technology. It’s the data.

Ireland produces plenty of information because farmers must record everything — herd numbers, breeding, pasture use, and emissions — for compliance or grants. But most of it is still entered manually, often late at night after the work is done. It’s enough for bureaucracy but not for decision-making. As one researcher put it, “The models are fine. It’s the inputs that fail us.”

That gap limits how much value technology can deliver. Farmers don’t resist digital tools; they simply have more urgent priorities. Until data entry becomes automatic — through sensors, voice input, or machine integration — the promise of digital farming will stay one step ahead of reality.

Common Pasture

The contrast between Ireland and Brazil agriculture is clear, but it doesn’t have to be a divide. In many ways, they could work better together than apart. Ireland brings depth in dairy systems, genetics, and pasture management. Brazil brings scale, sunshine, and the space to test new ideas. One refines, the other amplifies.

Ireland’s small, data-rich farms make it a perfect lab for precision and traceability tools. Brazil’s vast, diverse operations can stress-test those same innovations under real market pressure. Rather than competing, they could form a powerful cycle: ideas developed in Ireland, adapted and scaled in Brazil, and improved again through feedback and data.

The human bridge is already there. Nearly 100,000 Brazilians now live in Ireland — the second-largest foreign community in the country. They’ve found familiarity in the landscape, the humour, and the rhythm of life. Both cultures are welcoming, down-to-earth, and hardworking, with a shared sense that progress means pulling together.

In the end, the two farming worlds have more in common than their distance suggests. Both are proud agricultural nations learning to blend tradition with technology, resilience with renewal. Different horses, different courses — but grazing the same pasture.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.

Share

A Fragile Pact

For nearly two decades, the Soy Moratorium stood as a voluntary safeguard against Amazon deforestation. Major traders agreed not to buy soy grown on newly cleared forest land after 2008, a pact that reassured NGOs and international retailers. When Brazil’s competition regulator, CADE, recently suspended it on grounds of anti-competitive behavior, the delicate balance cracked.

What began as an industry commitment has become a legal and political battlefield. Supporters say it is one of the few effective tools to slow Amazon loss. Critics argue it creates “double law,” piling on requirements beyond Brazil’s Forest Code, which already obliges farmers to preserve large portions of native vegetation. Now, with appeals winding through the courts, the question now is not just about soy—it is about who gets to set the rules of Brazilian agriculture.

The debate intensified when twenty-five European supermarket chains—including Tesco, Sainsbury’s, and Aldi—wrote to ADM, Bunge, Cargill, Louis Dreyfus, and Cofco, demanding they continue zero-deforestation sourcing regardless of Brazil’s legal rulings. It was a blunt reminder that the real power in global supply chains often sits far from Brasília.

Above and Beyond

Under the Forest Code, Brazilian farmers are required to preserve anywhere from 20 to 80 percent of their land as native vegetation, depending on the biome. A farmer in Mato Grosso, for example, may legally clear part of his property for production while still remaining fully compliant with Brazilian law.

European buyers, however, have drawn a harder line. They insist on zero deforestation after 2008—no new clearing, regardless of whether it is legal under Brazil’s system. That is the sticking point. For long-established producers, already cleared before 2008, compliance is straightforward. For newer farmers who bought land more recently, the rule is effectively retroactive, blocking them from using portions of their own property that national law allows.

This widening gap between legal rights at home and market access abroad is at the heart of farmer frustration. Compliance is no longer about following Brazilian law; it’s about following European definitions of sustainability, even when they go further than national standards.

Clear Cutoff

In Mato Grosso, the world’s largest soy-producing state, the impact is particularly acute. Farmers who purchased land with the expectation of clearing it within the legal limits of the Forest Code now find themselves locked out of premium export markets if they do so. The moratorium’s cut-off date of 2008 penalizes newcomers, regardless of whether their clearing is lawful.

For established producers, this creates a perverse advantage. Their fields, cleared decades ago, pass certification checks with little effort. Younger farmers, who entered the sector later, face restrictions that make their business models unviable. The sense of injustice is palpable. Farmers argue that they are being asked to shoulder rules written in Brussels rather than Brasília, while Europe conveniently forgets its own past.

The hypocrisy stings. As many farmers point out, Europe cleared most of its forests decades ago, often with government subsidies. In my native Ireland, farmers were even paid in the 1970s to cut down trees and expand arable land. Today the same continent that once incentivized deforestation tells Brazil to keep its trees standing, this time without offering any financial incentive in return.

Trust and Trade

The moratorium debate is therefore less about legality and more about legitimacy. Brazil’s Forest Code already sets some of the strictest conservation standards anywhere. But for buyers, legality is not enough—they want assurances that extend beyond national law. For farmers, this feels like a breach of sovereignty and a denial of property rights.

The power imbalance is clear. Europe may account for a smaller share of Brazil’s soy exports than China, but its influence extends through the rules it writes into corporate supply chains. Traders, unwilling to lose access to European shelves, often enforce zero-deforestation standards across the board. In practice, European buyers are shaping land-use policy in Brazil, even without formal jurisdiction.

The result is a triangle of mistrust: farmers see hypocrisy, regulators see sovereignty under attack, and buyers see risk mitigation. The Soy Moratorium sits at the fault line of these competing perspectives.

Beyond the Moratorium

What happens next will shape not just soy, but the future of Brazil’s ag exports. One option is for Brazil to lean harder on Forest Code enforcement, showing the world that its own laws are credible and sufficient. Another is a continuation of buyer-led zero-deforestation commitments, effectively bypassing national law. The third, and most workable, is a hybrid: Brazil’s laws enforced through public systems, complemented by digital traceability that meets international demands at lower cost.

Brazil has already shown that productivity and preservation can coexist, through no-till farming, integrated crop-livestock systems, and regenerative practices across the Cerrado. The challenge is aligning that progress with the standards of skeptical buyers. If Brazil can prove, through transparent data rather than moral arguments, that legal production is also sustainable, it could shift the balance of trust.

The Soy Moratorium began as a voluntary pact. Today it is a flashpoint in a larger clash between sovereignty and market power. If Brazil can turn compliance into competitiveness, it won’t just defend its soy sector. It will redefine who really sets the rules of global agriculture.

Thanks for reading.

KFG

Kieran Finbar Gartlan is an Irish native with over 30 years experience living and working in Brazil. He is Managing Partner at The Yield Lab Latam, a leading venture capital firm investing in Agrifood and Climate Tech startups in Latin America.