If retrofitting agricultural processes with artificial intelligence were easy, or cheap, then everyone would have done it already. Though agriculture is estimated to be a $5 trillion industry, research suggests that output could be even higher with greater efficiency provided by AI. So what is stopping Ag operations from fully deploying artificial intelligence through sensors, machine learning, and the like?
A few seasoned veterans of the agriculture sector shared with us some insight into the challenges hampering further AI adoption. Here's what they said:
1. Brad Constantinescu, President and CTO of Stone Soup Tech
“Despite the advances described above, agriculture is still a very hands-on business with a lot of room for improvement. The major challenge with AI adoption, as in most industries, is bridging the gap between farmers and AI engineers. Farmers encounter a variety of problems that AI can fix. They may not consciously see the problem or they may not know it is solvable. Nobody thought of putting activity trackers on cows to detect early signs of illness – before a company (Cowlar) did just that and improved dairy production. Meanwhile, AI engineers know little about agriculture, the problems and the opportunities in this field.”
2. John McDonald, CEO at ClearObject
“Although AI has made inroads, agriculture remains a difficult sector to contain for the purpose of statistical quantification to realize AI’s value. According to Joseph Byrum in a 2017 blog post, even within a single field, conditions are constantly changing. Unpredictable weather, changes in soil quality, and crop disease can all affect statistical analysis. At the same time, ecosystems vary wildly from continent to continent, meaning things like seed and fertilizer program are rarely consistent globally. Therefore, AI and machine learning are still far from being able to predict critical outcomes in agriculture purely through the cognitive ability of machines.”
3. Jeff Klaumann, Chief Technology Officer at Internet of Things America
“Even though millions of people live and work in rural communities across America, telecommunications companies continue to invest in new business models and new technologies that improve connectivity options in urban, more densely populated areas. This common practice leaves many producers access disadvantaged. Precision agriculture and agtech innovations, combined with AI and data insights, track and improve productivity of operations for farmers and ranchers. More important, these technologies need connectivity in the field to function. With access to greater connectivity, relevant data is delivered in real time and can be sorted, analyzed and used to help ranchers and farmers make critical decisions that can improve their efficiency and, ultimately, their herd or crop productivity.”
4. Roger Royse, founder of the Royse AgTech Innovation Network
“There have been technical challenges – such as the lack of implementation of a rural broadband structure – that remain to be resolved, but more importantly, as in other industries, AI has promised more than it has delivered. We aren't quite there yet, but it is only a matter of time, as the use cases have been clearly identified.”
5. Kirk Haney, CEO of Radicle Growth
“AI is taking over the connected world, but the farm is not yet connected. Once the farm is connected, whether satellites, 5G, LoRA or something else, this is when AI will transform ag.”
6. Ash Madgavkar, Founder of Ceres Imaging
“The biggest challenge we see is making the insights that come from AI actionable. Farmers want to adopt AI but they aren’t going to waste their time if it’s not usable inthe field. AI powers new technology tools. And like all tools, if they are not useful, they won’t be used. Ag tech services need to focus heavily on making data actionable in a way that helps farmers prioritize issues in their fields in a more meaningful way.”
7. John Corbett, CEO of aWhere
“Agriculture is finally switching from being driven by cultural traditions to one where business intelligence guides behavior. The warming atmosphere can completely switch ecological conditions for diseases (i.e., fungus, molds) and farmers continuing old practices are likely to suffer catastrophic consequences.”
8. Gary Morgan, Director of MPT Innovation Group
“The major challenge in the broad adoption of AI in agriculture is the lack of simple solutions that seamlessly incorporate and embed AI in agriculture. The majority of farmers don’t have the time or digital skills experience to explore the AI solutions space by themselves. AI faces the same challenge as the war between AC and DC current did at the turn of the 19th century; it became more about the solutions that the technology-powered, rather than the technology itself.
AI solutions in agriculture will require new ontologies and common terminologies to be agreed upon globally. These new AI solutions will then have to be incorporated into existing and legacy infrastructure and systems that farmers already use (e.g. tractors, spreaders or Farm Management software), through improved APIs, in order to seamlessly incorporate and embed AI within agriculture.”
Have expert insights to add to this article?
Share your feedback and we'll consider adding it to the piece!ADD YOUR INSIGHTS