AI helps plants tell you when they’re thirsty

Have you ever joyfully gone out to your own backyard gardenwith freshly brewed coffee in hand, only to find your carefully tended plants and herbs wilted and dying? Was the soil too dry? did pests find their way? At times like these, some frustrated gardeners may wish their fickle ficus was just tell them what he needs A new Microsoft partnership project in the UK is trying to see if this concept can be demonstrated in real life.

Next year, the Royal Horticultural Society in England will show “smart garden,” which uses an artificial intelligence model to monitor the garden’s environment and inform gardeners when it needs tending. Visitors can ask questions of the AI-based garden. The model could then respond with phrases like “I need a little more water” or “I need a haircut” depending on the data received in the soil. In addition to the pleasant novelty of interacting with plants, the garden will provide visitors with a physical representation of the many ways AI and Internet of things technologies can be used to enhance resilience and further conservation efforts.

How AI Garden works

The garden’s technical components are the product of a partnership between AI startup Avande and Microsoft. Once open, the garden will rely on a network of sensors located throughout the garden that will measure environmental factors such as soil moisture, alkalinity and nutrient levels, as well as wind and forecasted rainfall. All this data is then sent to a specially designed artificial intelligence model located in a pavilion behind the garden. A model based on the open artificial intelligence service Microsoft Azure can analyze these factors and inform gardeners about optimal feeding, watering or pruning regimes.

Plant caretakers are not the only ones who can interact with the garden. Visitors to the Chelsea Flower Show will be able to approach the virtual platforms and ask the intelligence garden questions about its current ecological status. The AI ​​language module then interprets the data and communicates it to visitors in a coherent conversational dialogue. In theory, the insights provided by models like this could help gardeners allocate resources efficiently and avoid overwatering plants before rainfall, which could aid in sustainability efforts. These tools can be a welcome convenience for casual producers or businesses looking to cut utility costs. However, it is unclear whether even the best models can account for the anomalies and individual trends of plants with the same level of care as experienced gardeners.

Tom Massey, one of the garden’s two designers, described the project as “a unique opportunity to tackle some of the most challenging environmental challenges in horticulture”.

“I’m excited to see how artificial intelligence can help gardeners make a significant impact that will help us all garden and grow more sustainably,” Massey. it is stated in the press release.

AI assistance, not artificial intelligence

Messi and fellow designer Jae Ahn wanted to avoid the trap of making the AI ​​look too artificial. While sensors and algorithms play a critical role in maintaining the plants, the actual physical design of the space is left to the people. The result is a “forest garden concept” rich in biodiversity and greenery. The garden will reportedly live on as an accessible public space after the exhibition ends.

“Artificial intelligence didn’t design the garden, AI helps gardeners tend it,” Massey said in interview with The Independent.

While AI chatbots and image generators tend to get the most attention, more traditional AI models are already playing an important role in agriculture for resource management and sustainability. Farmers in the US are already using artificial intelligence to measure soil conditions to improve yields and protect against drought in a practice called “precision farming.” Everyday gardeners have already started practicing. Gardeners are reported to be using ChatGPT and other chatbots question the water or food requirements of certain plants, even though these models are still prone to it hallucinations of false statements. Sprat programs available also use artificial intelligence to help collectors identify unknown plants by comparing their photos to a large online database. But how past reports have shownsimply trusting artificial intelligence to tell you whether a plant or mushroom of unknown origin is safe to eat is unwise and can have dangerous consequences.