A groundbreaking encounter with a cyborg plant transformed Justin Mast’s path.
Several years prior, while delving into scientific papers, Mast discovered a study detailing a houseplant connected to a motorized platform. This plant, when needing additional light, would transmit electrical signals through its living cells. A connected control system detected these currents and autonomously moved the plant toward the light source.
“I thought, ‘What in the world is this?'” Mast shared with TechCrunch. “That moment initiated my deep dive into the subject.”
This led Mast to collaborate with Lee von Kraus, a neuroscientist with DARPA experience, to establish Electric Plant. For the last three years, the company has focused on gathering data to develop Bombadil, a large language model designed to decipher the faint electrical impulses within plant stems and communicate the plants’ requirements to cultivators.
The company’s founders envision their model eventually becoming a mechanism that enables humans to “converse” with plants. “Our current focus is on listening rather than speaking, which is the foundational stage,” explained Mast, co-founder and CEO of Electric Plant.
Electric Plant recently secured an $11.2 million seed funding round to launch its sensors and the Bombadil model. This round was spearheaded by Onto Ventures and Climactic, with additional contributions from Ponderosa, Endeavor Catalyst, Ludlow, Michigan Rise, The Fund, Schox VC, and Dug Song.
Mast noted that even in optimal greenhouse conditions, cultivators lack comprehensive insight into their plants’ specific needs. Historically, agricultural professionals have depended on less precise instruments that respond to plant requests for water, nourishment, or light. While existing tools such as soil probes, handheld spectroscopes, and satellite imaging are functional, Mast believes they fall short of ideal.
He remarked, “While current technology is advanced, its capacity to assess a plant’s internal state lacks refinement. Much of this equipment still trails significantly behind a plant’s immediate requirements.”
Nevertheless, Electric Plant commenced its development by utilizing existing devices to train its Bombadil model, which can now decipher electrical signals from approximately 125 distinct plant species. Mast stated, “We are confident that our model will outperform much of the current equipment.”
Furthermore, the startup’s sensors offer a considerable cost advantage over current tools, which often run into thousands of dollars. He noted, “Our solution costs mere tens of dollars.”
These sensors, roughly the dimensions of an iPhone folded in half, are affixed to plants with stainless steel screws. To detect the minute microvolt fluctuations inside the plant, the screws need to pierce the cambium, which is the living tissue layer directly beneath the bark. Information gathered by these sensors is then transmitted wirelessly to a base station using a low-power network.
According to Mast, Electric Plant requires only a few sensors per acre to achieve precise readings. The collected data is processed by Bombadil in the cloud, and the insights are then returned to the client, enabling them to optimize aspects like watering and fertilization schedules. Mast asserts that this system can enhance crop yields and simultaneously decrease water consumption by at least 20%.
While Mast refrained from revealing Electric Plant’s initial clientele, potential applications in the agricultural and forestry sectors are readily apparent.
Mast mentioned that Bombadil is proficient with over a hundred species, encompassing a wide array of plant types, including ferns, redwoods, tomatoes, and dahlias.
“This represents the most comprehensive model ever developed for interpreting biological electrical signals,” he declared, adding, “Nothing comparable has been achieved in the animal kingdom.”