Reading plants' electrical pulses with AI shows low-cost sensors can provide a fifth of irrigation water

The story begins with a unusual experiment: a houseplant whose living tissue was connected to a control circuit and a movable base that nudged it toward a light source whenever electrical signals indicating its need for energy were present. This scientific experiment prompted Justin Mast, together with neuroscientist Lee von Kraus, who previously worked with the Defense Advanced Research Projects Agency (DARPA), to found the company Electric Plant to decode the silent biological signals inside plants. Over three years, the team gathered physiological data to train a large language model called Bompadel, capable of translating the minute electric pulses flowing through plant stems into understandable data that precisely determines crop needs for farmers.

Farmers and forest managers have relied for decades on approximate traditional tools to read field requirements for water, nutrients and light, using soil probes, portable spectrometers, and satellite imaging. Although those technologies have advanced, they remain external monitoring tools that lag behind the organism’s moment-to-moment needs. The essential difference offered by the startup is that it developed sensors the size of a folded iPhone that attach directly to a plant’s stem with stainless-steel screws. These screws penetrate the cambium, the active living tissue layer directly beneath the bark, to capture instantaneous microvolt-scale electrical changes and transmit them via a low-power wireless network to an intermediate station and then to the cloud for processing by the intelligent model.

Direct listening to plant cells breaks the dominance of expensive field sensors.

The issue does not stop at inventing a physical sensor; it also extends to a unprecedented biological coverage, as the model has been trained to read electrical signals from roughly 125 different plant species, ranging from tomatoes and dahlias to ferns and giant redwood trees, making it the most diverse model for handling biological signals, according to its founders. Moreover, there is a decisive economic gap: while spectroscopic and physical analysis instruments cost thousands of dollars, these sensors cost only a few dozen dollars, and two or three units per acre are enough to provide reliable results that enable farmers to recalibrate irrigation and fertilization schedules with a precision not previously available.

Electric Plant raised a seed financing round of $11.2 million led by Unto Ventures and ClaimTech, with participation from prominent investors including Endeavor Catalyst, Loodlo, Michigan Rise, The Fund, Shocks VC and Doug Song. The funding is aimed at commercializing the sensors and the Bumpadel model in the agriculture and forestry sectors. Company data and statements from its chief executive indicate that the system can increase crop yields while reducing water use by 20 percent or more, a figure that represents an exceptional operational impact in intensive agricultural production environments and greenhouses.

This shift in agricultural sensing physics directly touches the core of operational challenges in the Gulf, Egypt and the Levant, where the cost of desalinating and pumping water constitutes the largest burden on agricultural budgets and protected projects. The prevailing reliance in desert farms today rests on soil moisture or satellite images that only capture crop stress after it occurs and the leaves change colour; this means that agricultural decisions always arrive as a delayed reaction that consumes additional water to compensate. Switching to sensors that read water stress from within living tissue at a cost of only a few dozen dollars per unit gives greenhouse managers and desert reclamation farm operators in the region a unprecedented ability to schedule drip irrigation, turning a twenty-percent saving in water consumption into an immediate and direct reduction in energy and operating bills, as well as cutting waste of imported fertilizers.

The real value of farms and production projects today goes beyond taking generic satellite images of fields; it lies in probing the living tissue itself before signs of drought or nutrient deficiency appear on the crop. A preemptive response to an electrical signal lasting seconds may protect crops worth millions of dollars and make the difference between a losing season and a profitable farm.