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

If you're involved in agriculture or forestry, expect to use cheaper, more precise tools to monitor plant health and resource needs.

Electric Plant, a startup co-founded by Justin Mast and Lee von Kraus, has developed an AI model named Bombadil to interpret electrical signals in plants. The model, trained over three years using data from over 125 species, aims to decipher plant needs related to light, water, and nutrients. These electrical signals, previously unharnessed for practical applications, are detected by low-cost sensors, costing tens of dollars, which attach to a plant and transmit data wirelessly. The data is processed in the cloud by Bombadil, providing growers with real-time insights to optimize watering and fertilizer schedules. Mast claims this system can increase crop yields by at least 20% while simultaneously reducing water consumption. The company recently secured $11.2 million in seed funding, co-led by Onto Ventures and Climactic, to bring its technology to market, potentially revolutionizing agriculture and forestry by offering unprecedented visibility into plant well-being.

The development of Bombadil signifies a notable advancement in bio-integrated technology, moving beyond traditional external sensing methods to directly interpret a plant's internal electrical communication. This approach, validated across over 125 species, suggests a potential paradigm shift in how we understand and interact with botanical life, analogous to breakthroughs in understanding animal signals.

This innovation could democratize advanced plant monitoring, making sophisticated insights accessible to a broader range of growers due to the significantly lower cost of Electric Plant's sensors compared to existing technologies. The potential for yield increases and water use reduction positions this technology as a key enabler for sustainable agricultural practices and a valuable tool in climate change adaptation efforts.