Plants can speak using electricity, so this startup built an AI model to decipher their language

Justin Mast says a single weird research paper sent him down the rabbit hole. It described a houseplant wired to a motorized stand, with electrical signals in its tissues guiding the plant toward light. That idea led him to cofound Electric Plant with neuroscientist Lee von Kraus, who has worked with DARPA.

For the past three years, the startup has been collecting data for Bombadil, a large language model built to interpret the electrical signals running through a plant’s stem and translate them into something growers can use. Electric Plant says the system already covers about 125 species, from ferns and redwoods to tomatoes and dahlias. Mast calls it the most diverse model ever built on biological electrical signals.

The company just raised $11.2 million in seed funding, led by Onto Ventures and Climactic, with backing from Ponderosa, Endeavor Catalyst, Ludlow, Michigan Rise, The Fund, Schox VC, and Dug Song. The pitch is straightforward: plants are already sending signals, but farmers usually see the problem late and through blunt tools like soil probes, handheld spectroscopes, and satellite imagery.

Electric Plant wants to sit closer to the source. Its sensors are about the size of an iPhone folded lengthwise, attach with stainless steel screws, and have to reach the cambium, the living layer just under the bark. The data goes over a low-power wireless network to a base station, then into the cloud, where Bombadil interprets it and sends results back so growers can adjust watering or fertilizer. Mast says the sensors cost tens of dollars, not thousands, and that the system could boost yields while cutting water use by 20% or more.