Startups

Electric Plant's AI Model Decodes Plant Electrical Signals to Guide Farm Management

A startup built Bombadil, an LLM trained to interpret electrical signals from plants, to help growers optimize crop health in real time.

Last verified:

Electric Plant’s $11.2M Seed Round to Scale Plant-Electrical-Signal Interpretation

Electric Plant, co-founded by CEO Justin Mast and neuroscientist Lee von Kraus, secured $11.2 million in seed funding led by Onto Ventures and Climactic, with participation from Ponderosa, Endeavor Catalyst, Ludlow, Michigan Rise, The Fund, Schox VC, and Dug Song. The company is bringing to market a combination of low-cost biosensors and Bombadil, a large language model designed to translate electrical microvolt signals from plant tissues into real-time agronomic guidance. The funding allows Electric Plant to scale from a three-year prototype phase toward commercial deployment in greenhouses and field agriculture.

How Bombadil Interprets Plant Electrical Communication

Plants emit measurable electrical signals as they respond to environmental stressors—drought, nutrient deficiency, excess light, and disease. According to TechCrunch, Electric Plant’s approach involves attaching stainless steel screws to the cambium layer (the living tissue beneath plant bark) to sense these microvolt fluctuations. Data flows wirelessly to a base station, then to Bombadil running in the cloud, which outputs actionable intelligence about the plant’s water, nutrient, and light requirements. The model was trained across approximately 125 species using conventional monitoring equipment—soil probes, handheld spectroscopes, and satellite imagery—as ground truth, allowing it to learn the relationship between electrical signatures and observable plant stress. Mast told TechCrunch that the model will eventually surpass the accuracy of the legacy tools used to train it.

Economics and Deployment Density

A major differentiator is cost. TechCrunch reports that Electric Plant’s sensors cost tens of dollars per unit, compared to thousands for equivalent handheld or fixed-station alternatives. The company also claims deployment efficiency: only a couple of sensors per acre are sufficient for accurate field-level results, reducing the total cost of instrumentation relative to traditional grid-based monitoring. This efficiency and price point could enable adoption across small and mid-sized farms where equipment budgets are constrained.

Why This Matters

The agricultural sector has historically lacked real-time visibility into plant physiology during the critical growing season. Growers rely on reactive water schedules, post-hoc soil testing, and visual symptom detection—all of which introduce latency. If Bombadil’s electrical-signal interpretation proves reliable across diverse crops and environmental conditions in field trials, it could shift the economics of precision agriculture toward smaller operations and reduce water and fertilizer waste at scale. The challenge ahead is demonstrating that the model generalizes beyond the 125 species in its training set and that sensor reliability holds across years of field exposure. Investors’ confidence suggests they believe Electric Plant has solved a meaningfully hard problem—translating a novel biosignal modality into actionable decisions without requiring domain expertise from the end user.

Frequently Asked Questions

How does Electric Plant measure what plants 'need'?

Stainless steel screws penetrate the cambium layer under plant bark, sensing microvolt electrical changes. A base station collects this data wirelessly and sends it to Bombadil in the cloud for interpretation.

What is Bombadil trained on?

According to TechCrunch, Electric Plant trained Bombadil using conventional monitoring equipment (soil probes, handheld spectroscopes, satellite imagery) as ground truth, across approximately 125 plant species.

How many sensors does a grower need?

Electric Plant reports that only a couple of sensors per acre are required for accurate results.

How much cheaper are Electric Plant's sensors than alternatives?

The company's sensors cost tens of dollars; competing tools like handheld spectroscopes and soil probes typically cost thousands of dollars.

#ai-for-agriculture #large-language-models #biosensing #funding