The combine drives itself across the field — GPS autopilot holding the line to within a couple of centimeters, the planter metering seed by the meter, sensors under the header counting every kilogram of grain that drops into the tank. The farmer sits in the cab more as an observer than a driver. It's efficient, it's the future of agriculture writers imagined decades ago. There's just one question that rarely gets asked out loud: where do all those numbers go once they leave the field — and whose are they when they land?
The answer usually doesn't favor the person who planted the crop.
A Combine That Counts More Than Grain
Modern precision agriculture is not just a GPS steering wheel. Every major machine — tractor, planter, header — now carries a modem and a suite of sensors writing, in real time: soil type and moisture at every square meter, yield at a resolution of a few meters, fuel burn, engine health, exact tracks of every pass across seasons. Satellite imagery, drone scans, and weather data get stitched on top, building a digital map of the farm detailed enough to forecast next year's yield more precisely than the farmer standing in it can.
John Deere's Operations Center is the largest of these platforms, but not the only one: Bayer runs Climate FieldView, Corteva has its own stack built partly on Granular, which it acquired years ago. All three work the same way: the farmer's own equipment continuously streams telemetry to the manufacturer's servers, and in return the farmer gets tidy dashboards, planting recommendations, and a promise of a better yield. The deal looks fair right up to the clause about who actually owns the data it collects.
Whose Field, According to the Fine Print
The terms of service for these platforms usually name the farmer as the "owner" of their field data. But that's almost always followed by a broad license letting the company aggregate, analyze, de-identify, and use that data to "improve the product," and share insights with partners — seed companies, insurers, fertilizer suppliers. "De-identified" sounds safe, but a field's yield is tied to its GPS coordinates, which trace right back to the owner if anyone bothers to run the match.
Farm groups — in the U.S., mainly the American Farm Bureau Federation and the National Farmers Union — have spent years pushing agtech companies for real data portability: the ability for a farmer to pull their own data and take it to another provider without starting the field's history from zero. There's been progress, but it's uneven: a decade of yield history built up on one platform effectively locks a farm into that manufacturer's ecosystem, because moving that history to a competitor's system is either technically blocked or made cumbersome enough that nobody bothers.
The Fight Over Fixing Your Own Tractor
Running alongside this is a simpler, more visceral version of the same fight: the right to repair your own machine. A modern John Deere tractor is so loaded with proprietary software that even swapping a basic sensor requires a diagnostic code only an authorized dealer can issue. A farmer who paid several hundred thousand dollars for the machine can't fix a breakdown mid-harvest — they have to wait for a service technician or haul the equipment across the state.
In 2023, Deere signed a memorandum with the American Farm Bureau Federation promising to give farmers access to diagnostic tools and documentation. The promise turned out to be only partly kept: in January 2025, the U.S. Federal Trade Commission, joined by several states, sued Deere & Company, alleging the company had artificially maintained a monopoly over repairs of its own equipment. Colorado had become the first state to pass a right-to-repair law specifically for farm equipment the year before. This isn't a separate story from the data fight — it's the same logic of ownership, inverted: both the code in the engine and the numbers in the cloud belong to the manufacturer, even while sitting on your land and grown from your field.
Who's Counting the Harvest Before You Do
The data collected here isn't just convenience. It's competitive advantage, and it's concentrating in very few hands. After buying Monsanto in 2018 for roughly $63 billion, Bayer didn't just get a bank of seed patents — it got Climate FieldView, a platform holding years of yield history across millions of fields worldwide. That kind of dataset lets a company recommend, with real precision, how much fertilizer and which specific seed to buy for that exact field — a recommendation that, conveniently, is almost always its own product.
Further up the chain sit the global grain traders known as the "ABCD group" — ADM, Bunge, Cargill, and (Louis) Dreyfus — which, by various estimates, handle the large majority of the world's grain trade. An early, accurate read on regional yields is a direct market edge for contracting and hedging: whoever knows the state of the fields before the market does, trades before the market does. The farmer whose own data fed that forecast finds out the price last.
Increasingly, it's not just processors chasing that data but investors. Institutional funds and large private buyers have been acquiring U.S. farmland and land abroad — business press coverage points to sizable acreage changing hands this way — and how predictable a field's future yield looks, reconstructed from its digital history, feeds directly into what buyers are willing to pay for it. The average American farmer, per the agricultural census, is now over fifty-five; land is changing hands faster than most people assume, and increasingly the data about it is worth more than this year's crop.
Our record: in ancient Egypt, state power rested partly on measurement — the annual Nile flood was tracked with a nilometer, the harvest converted into tax, and whoever held the accurate measure held real leverage over the people who worked the land. Ma'at, in its engineering sense rather than its moral one, is precisely about the true measure and its fair distribution: measure, weigh, return a share to the one who produced it. Today's nilometer is a satellite and a sensor bolted under a header, and the temple that keeps the measure is an agtech corporation's server. The harvest is measured more precisely than at any point in history, but the measure flows one way — upward, into someone else's property, with no share returned to the land and labor that generated it. That's Isfet in its most mundane form — not a conspiracy, just terms of service nobody read.
The Lever the Farmer Still Has
The agricultural data market isn't as locked in as it looks. Data portability — the ability to export a field's full history and take it to another provider or an independent system — is already part of farm-group advocacy in several countries, and where it's actually written into law, it genuinely breaks the single-platform lock-in. The right to repair, won in Colorado and now being argued in the FTC's suit, works on the same principle: owning the machine should mean controlling its code, not renting that control from the manufacturer one click at a time. Independent, cooperative agricultural data platforms are also growing — places where farmers jointly own the infrastructure and decide themselves who gets the data and on what terms, instead of signing that right away the moment the equipment powers on.
The land doesn't stop being yours just because someone else's algorithm now measures what it grows. But it risks becoming someone else's on paper before it ever shows up on a deed — if the only party that actually knows what the land is worth, and what it will yield next, is sitting not in the combine's cab but in a data center a thousand miles away.