Why AI in Farming Must Bow to Biology, Not the Other Way Around

Jan de Vries · 2026-08-17

As agriculture wrestles with more volatile weather, stricter rules, and pickier buyers, the big question isn't whether AI belongs on the farm. It's whether AI can actually work on biology's terms. Spoiler: it has to. Munich-based AgriTech startup Greenda, founded in 2024, is building a data and decision layer for farmer organizations. The goal is simple: help co-ops turn scattered field knowledge into timely, independent, and actionable advice. In this interview, Greenda's Chief Scientific Officer and environmental microbiologist Sabrina Pittroff gets real about who owns farm data, why farmers and agronomists must stay in the driver's seat, and why the next wave of AgriTech should support ecosystems instead of trying to overpower them. ### The Push From Lab to Field Pittroff didn't follow the typical academic path. "I never quite fit the 'publish or perish' mold," she says. At university, she loved digging into curiosity-driven projects—the kind where you can sit with a mechanism long enough to truly understand it. But the frustration came after. "You can write a paper pointing clearly toward an application and still have no guarantee anyone picks it up," she explains. "That uncertainty, that slowness—that's where my impatience lived." So she turned to entrepreneurship. It let her use the same research skills to build a solution, not just describe one. The hard part, she admits, is the gap between research and a real product. "You need scientific integrity and the ability to develop, communicate, and sell." That's where Chadi Nemr, Greenda's CEO, came in. "The Greenda moment, for me, was recognizing he had the vision and experience to lead the parts of building a company that I didn't—and that together we could close that translation gap." ### Biology Always Wins Pittroff is blunt about her industry. "Agriculture is a strange industry, and at times a corrupt one." But she's drawn to it for one simple reason: > "Underneath every business model, every regulation, every market dynamic, there is biology—and biology has the final say. You can build around it, but you can't override it." She finds hope in the fact that nature has already designed most of the solutions we're trying to invent. "Most of our work, honestly, is getting out of our own way and showing data to support or advocate for a healthy and active ecosystem." ### Who Really Controls Farm Data? The honest answer? Nobody—because most of it isn't collected at all. "It lives in paperwork, in unformatted spreadsheets, in the heads of farmers and agronomists who are themselves remarkable data-collection machines," Pittroff says. Walk a field with a good agronomist and you'll see decades of pattern recognition that no software has ever captured. But here's the catch: humans can't see patterns the way machines can—not across plots, seasons, and regions. That's the real gap. And when you look at the small fraction of farm data that has been digitized, it tends to flow upward in the value chain, not back to the farmer who generated it. ### Flipping the Power Dynamic Pittroff doesn't mince words about the imbalance. "Farmers have historically been the last in a long chain of consequences—last to capture margin, last to set terms, last to access their own information." That's why Greenda exists. "Greenda is built on the belief that farmers should benefit from the data their land produces." Bringing even basic data into one structured place gives farmers leverage they've never had before. For agronomists and farmers, this means AI isn't a replacement—it's a tool that amplifies their expertise. The future of AgriTech isn't about machines taking over. It's about machines finally listening to the people who know the land best, and to the biology that ultimately calls the shots. ### The Bottom Line If you're in agtech, the lesson is clear: build for the ecosystem, not against it. The winners will be the ones who respect the natural order and put data back into the hands of the people who create it.