New technology can help farmers and food producers do their jobs more effectively and efficiently, a potential benefit for those working the 150-million acres of farmland across Canada.Getty Images
A campfire might seem like an unusual setting for a discussion on artificial intelligence (AI) and farming, but it’s where the idea for BETSY the AI Ranch Hand was born in 2019.
“My cousin, a farmer in Saskatchewan, asked us if we could ID his cows,” recalls Mokah Shmigelsky, founder and CEO of OneCup, the company that developed BETSY. “My husband Jeffrey has a master’s degree in artificial intelligence and computer vision. He said, ‘Yes, I bet we could do that. There’s no reason you couldn’t.’”
BETSY uses cameras and data to help ranchers and farmers identify their animals quickly, keep them in good health, know when they are in distress, and send signs of labour during calving season. All of this is done with a high level of accuracy, meaning fewer animals lost to illness and harsh elements. The technology is now set up on more than 200 farms across Canada and the northern United States.
AI technology such as BETSY can help farmers and food producers do their jobs more effectively and efficiently, a potential benefit for those working the 150-million acres of farmland across Canada. It can also improve the division of labour, aiding in more menial tasks, saving money, and alleviating worker shortages in the industry.
But farming is a traditional sector and a big cultural shift is required to get people on board with new technology in their daily operations.
“A lot of our selling proposition has been in the education of producers in how this technology can be used, how it is installed. But we’ve designed it to be as easy as possible,” says Ms. Shmigelsky, adding that many farmers have some sort of camera system installed and OneCup’s cloud-based technology simply piggybacks on those systems.
Potential cost savings is a big part of trying to attract more users, Ms. Shmigelsky adds. “If you save one calf in the process of a calving season, you’ve essentially paid for the system itself. That doesn’t even take into account the time savings and the ability for these producers to get a good night’s sleep.”
The margins in livestock farming are often lower than that of crop farming, so it makes sense that cost would be a barrier for livestock farmers, says Sean Thompson, director of the Technology Access Centre for Livestock Production at Olds College of Agriculture and Technology in Olds, Alta.
“On the livestock side, mainly beef cattle and sheep, adoption is slower partly because of financials and lower margins, but also that social aspect, not wanting to rely on a camera system to tell you [about your herd] or having that lack of confidence in some of these technologies to actually make decisions on your farm.”
The challenges AI could solve should make it more attractive to those in the agriculture business, particularly when it comes to cost and labour. Canada is already experiencing a labour shortage in the farming industry, and the domestic labour gap is expected to increase to more than 100,000 workers by 2030, according to a recent report by the Canadian Agricultural Human Resources Council.
“Animal husbandry is not something everyone can jump into,” Mr. Thompson says. “If you can rely on computer vision and AI to help with that, not to take over completely but to help, that really will help with the bottom line and a lot of operations.”
The story is different on the crops side of farming – particularly controlled-environment farming or vertical farming – where AI is taking all kinds of forms, from yield predictions to tracking water intake, creating huge value for producers.
At GoodLeaf Farms, the only vertical farming operation with a national footprint, the average time from seed to packaging for lettuce, spinach, kale and other greens is 18 days (the growth time on a typical open-field farm is 45 to 60 days). The popular microgreens go from seed to package in eight days.
This efficiency is because of AI, says Barry Murchie, president and chief financial officer of GoodLeaf Farms. “We take over 100,000 data points a day from our farms, and it just gives us the ability to constantly be refining and improving every aspect of what we do.”
AI is critical to this type of farming, which Mr. Murchie describes as a “plant factory,” where salad greens are grown in a Costco-like warehouse with no natural light and where only 5 per cent of the amount of open-field farming water is used. Cameras and data are used to track plant growth and requirements to ensure optimal production.
“We have predictable outcomes every time we plant,” says Mr. Murchie. “We’re achieving more nutrient-dense plants with a longer shelf life. They look better. They taste better.”
To be skeptical of this type of AI-assisted food production is becoming increasingly difficult as the world deals with challenges such as water instability, pesticide use, food insecurity, and deterioration of plants going through the supply chain, Mr. Murchie explains.
“These aren’t some far-off concerns that may or may not happen. These are acute problems that are happening now. If we do not find alternatives that answer and bring solutions to those problems, we’re going to have issues. As a result, we’ve been on this journey to bring a viable alternative.”