| Alex Kirkpatrick: Hello. I’m Alex Kirkpatrick from the Center for Sustaining Agriculture and Natural Resources at Washington State University. This video explores a classic societal-level model of technology adoption, Diffusion of Innovations Theory, and what this might tell us about the likely AI adoption process in agriculture. Then, we’ll dig down onto the individual level, the united theory of acceptance and use of technology, and what this might tell us. Using behavioral science and psychological theory can really amplify your effectiveness as a communicator of innovation like AI. Let’s explore. | Text on screen, “Alex Kirkpatrick, PhD.” Alex speaks in front of a multicolored screen. |
| [ Music ] | A circular combine harvester cuts through a field of golden wheat. As it does, a woman controls it via a touch pad. Then, in a large greenhouse, liquid sprays from the tube that stretches across the vegetables growing below it. In a field, a woman uses a tablet to take a photo of a soybean plant. Then, a human-controlled harvester drives over and cuts wheat. The logos for “Western SARE,” “Washington State University,” “AG AID institute” and “National agricultural food and aggregation.” Text on screen, “Technology adoption. How and why people might choose to adopt AI innovations.” |
| USDA Speaker: Today, agriculture is going far beyond nature to produce new miracles for an even better, more abundant life. Scientists, working to improve plant quality through breeding, have given potatoes shallow eyes so they’re easier to peel, and they have also improved storage methods. Modern communications are helping this farmer sell his lettuce. This is just one of 200 perishable products, which move to market when and where needed, thanks to an intricate market information system. | Archival footage of an orange grove. A man fills a tin bucket with water, while another writes on a pad. Text on screen. “USDA, 1960. Miracles from agriculture.” Attached to a harvester, a conveyor belt delivers potatoes into a wooden container. From his car, a farmer talks on a phone and workers pick cabbages in a field. |
| Alex Kirkpatrick: Agriculture is dependent on technological change in order to keep a hungry world fed. Farmers, land managers, and agricultural operations of all kinds are presented with countless potential innovations that might help enhance efficiency, sustainability, and economic growth. | As Alex speaks, more footage shows the plants being loaded by hand onto a conveyor belt being pulled by a harvester. Then, hundreds of carrots drop into a container, and cows stand in pens. |
| Choosing between them, whether to adopt or ignore, can sometimes be the difference between success and failure. You can help inform these decisions by knowing more about the kind of information that’s most pertinent and useful to your audience. First, let’s look at the classic societal-scale theory, Diffusion of Innovations. An innovation can be a thought, a practice, a product, or a theory. The way these innovations spread from one category of user to the next has been found to follow a now very famous s-shaped curve. [ Music ] In theory, if you could establish what percentage of a system is using an innovation, you know which group is currently deciding to adopt or not next. Only about 2.5% of people are innovators. Innovators are eager to try new ideas and technologies, often at a higher risk. They’re well connected, financially resourced, and have higher risk tolerance. They pay attention to media more because they have nobody else to learn from because they’re usually the first in their social groups to find out about new things and experiment with them. These are the techies who likely adopt and experiment with new technologies regardless of their potential to become widespread. About 1% of Americans drive electric vehicles, for example. About 2% own any cryptocurrency. So we’re still at the innovation stages of these technologies. | Alex speaks in a studio. Text on screen. “Tech adoption on the societal scale.” A line graph shows an adoption scale rising from almost zero to almost “100 percent.” A block appears beneath the start of the curved line with the text, “innovators,” within. |
| In an ag-AI context, innovators are likely to be the larger operations who can afford to take a risk. Large public firms, like John Deere, Monsanto, and Corteva, can invest heavily in R&D because they can afford it. University farms also help to lead the way by integrating the latest research and experimental technologies. Innovators help to reduce uncertainty for early adopters by demonstrating the viability of new technologies, providing critical feedback, and normalizing AI adoption within their wider communities. We can engage innovators by showcasing cutting-edge research and developments. They’re likely interested in the technical details more so than later groups. Jargon isn’t such a barrier to these folks, which isn’t true of nearly everyone else who comes after. So providing access to technical papers and technical people is probably a good idea. You likely know who these innovators are among your stakeholders and within your own innovation networks. A great way to engage them in a way that appeals is through offering access to beta tests and pilot programs, hands-on field days at university farms, et cetera. Being more technically minded and switched on to the cutting edge, many innovators will likely be influenced by niche specialist media in their field, trade journals, and associated social media. Scientists and engineers are generally quite good at engaging this tiny minority of adopters because, very often, they’re innovators too. They talk their language. They often share similar worldviews or education levels, which can expedite communication. It’s important to still emphasize certain risks to innovators, as they can be somewhat risk-blind and overly venturesome. | As Alex speaks, a heading reads, “Innovators are receptive to technical communications.” This is followed by bullet points. “Technical details and specs.” “Beta tests and trials,” “networking opportunities.” “Specialist media.” “Encourage face time with scientists and engineers.” The curved line graph. |
| But what about the next group on the curve? The early adopters. This group has a high degree of opinion leadership. | The line graph reappears. Beside the “2.5 percent” innovators block, another block reads, “Early adopters at 13.5 percent.” |
| Early adopters are generally more integrated into the local community than innovators, more socially influential and networked. They’re likely more discerning in their adoption choices than innovators. In the context of agricultural AI, early adopters are typically progressive farmers, agronomists, and extension agents who are open to trying new AI tools after seeing successful case studies or hearing positive feedback from innovators. It’s important to show early adopters how innovations can be leveraged for competitive advantage and leadership within their business and social circles. We’re at that point with ag AI when the early adopters are becoming the next key audience now deliberating the risks and benefits of AI innovations. They’ll want to be shown evidence of return on investment. They’ll benefit from testimonials and case studies from whoever the influential innovators are to them. Maybe that’s a university or a particular researcher or an innovative firm that they trust and respect. Maybe it’s you. When in doubt, ask. Early adopters look for both functionality and proof of concept before adopting a new technology. Webinars, workshops, practical guidance on implementation are all useful for early adopters when making their decisions. Before widespread adoption, mass media become really central to bringing awareness of innovations to these influential early adopters. So keep your eye on how AI and ag AI, in particular, is treated in mass media for clues as to what these early adopters might be thinking about AI. | Alex reappears in front of the multicolored background. Beside Alex, a heading reads, “Early adopters seek proof before taking a risk.” This is followed by bullet points. “Practical ROI evidence.” “Peer recommendations.” “Proof of concept.” “Actionable guidance.” “Mass media influence.” |
| Next, we cross what we call the chasm. This marks a transition from the socially influential, performance-seeking early adopters to the more pragmatic and skeptical early majority. We call this the chasm to signify the difficulty of spreading a new idea to the early majority relative to the early adopters and innovators. Lots of innovations, like MiniDiscs, Betamax, Google Glasses, the Segway, find some traction among techie and venturesome innovators and advantage-seeking early adopters, but then many innovations stall. Ultimately, what will push ag-AI innovations over that gap will be how well AI performs in the hands of the early adopters. | The curved line graph reappears. A new block appears beside the Early Adopters block. Between them, a red dotted line rises to the top of the graph. The block reads, “Early majority, 34 percent.” |
| The early majority are less willing to take risks than early adopters and wait for clear evidence that a technology is reliable, cost-effective, and easy enough to integrate. In that sense, social influence becomes central to their decision. They want to see hard evidence of specific and widespread successes. They want to see ag AI in action through multiple channels, ranging from the interpersonal to the mass and computer mediated. As time goes on, the yet-to-adopt groups are naturally more cautious, risk-averse, and less inclined to waste time or resources on things that potentially might not work for them, or they just might entail too much cognitive effort or learning. In that sense, translation of jargon and complex technical specifications and language is key. The early majority is still willing to take a manageable chance, but not on something that’s too complex to understand. To appeal to these folks, we could use practical demonstrations, and that goes for any AI innovation. We wouldn’t want to get too fixated on the technical aspects or technical knowledge, though. Communication to this audience should emphasize the practical, personal risks and benefits of adopting the technology, including productivity enhancements and ease of use. | Beside Alex, a heading reads, “Early majority want to see reliable wins.” This is followed by bullet points. “Proof of widespread success.” “Hard evidence of reliability.” “Proof of risk mitigation and tolerable threat.” “Less tolerant of jargon.” “Show, don’t tell.” “Emphasize personal utility and ease of use.” |
| At this point, the curve sort of takes care of itself, theoretically. We’ve hit a critical mass almost. The late majority are the sceptics, about a third of any overall population. The late majority adopts a new technology mainly because of the economic necessity or insurmountable peer pressure. They’re typically less financially secure than early adopters, less able to wager on potentially faulty ideas or technologies. Take the internet in the mid-1990s. Not everybody had it or needed it at home, necessarily. But now, it’s difficult not to be online. At this point, a technology is almost a fact of life, a given, a must-have. | The curved line graph reappears. The next block reads, “Late Majority at 34 percent.” This rises to where the line begins plateau. |
| Here’s what the late majority might need in terms of communication. Address doubts explicitly, focusing on simplicity, support, and monetary value. This is where your ability to contextualize an innovation or address controversy and ideological concerns really come into play. They may want assurances, warranties, or explicit technical support. Overall, they’re looking for proof of social normalcy, and the ag AI is the safe bet. We certainly can’t provide this kind of evidence just yet. | Alex reappears. Beside Alex, a heading reads, “Late majority need strong assurances.” This is followed by bullet points. “Addressing doubt and uncertainty.” “Address controversy, fact versus fiction.” “Assurance, insurance, warranty and facilitating conditions.” “Perceived normalcy.” |
| And finally come the laggards. Highly skeptical of change, highly traditional. They adopt an innovation only when it reaches this critical mass and they have no other choice, or they never adopt. About 75% of Americans use some form of social media. So, while some, perhaps most of the late majority have adopted, there isn’t that leap to the laggards, and probably won’t be. Similarly, there are those agricultural operations that are still highly traditional. There isn’t a lot you can do here as a communicator, really. Ultimately, they will be persuaded by absolute necessity or nothing at all. At this stage, in 2025, AI is not a necessity to agriculture and it may never be. | The curved line graph reappears. Beside the next block reads, “Laggards, 16 percent.” |
| Overall, you can inform your communication strategy by first establishing what percentage of people in a particular group have already adopted an innovation. You can then look at the next group to potentially adopt and infer what types of information they might need to make an informed decision. If you know your audience well, you probably know what kind of adopter group they naturally fall into, and so can adapt the resources you offer and the messages you tailor accordingly. But other social scientific models can tell us more about how the individual comes to an adoption decision about IT, like artificial intelligence. | In the studio, Alex speaks in front of the multicolored screen. |
| [ Music ] The Unified Theory of Acceptance and Use of Technology provides a robust framework for understanding how the individual makes decisions about adopting new information technologies, like AI and ag tech. The dependent variable is usage behavior, which is naturally informed by intent. Four key variables influence adoption intention. They are all subjective and perceptual; therefore, you can influence their strength in potential users. First is performance expectancy, which is the degree to which an individual believes that using the technology will help them to achieve gains in personal performance or other things that are important to them, like more time off work or a shorter commute. The question to help answer for your audience is: How will this technology improve them? Effort expectancy is the ease of use associated with the technology. All people value relief from cognitive load or mental stress. If people perceive a high barrier to entry, too much learning, too much intricacy and dense user manuals, they’re less likely to adopt. The question to help answer here is: How easy is this innovation to understand, use, and integrate? Then, there’s perhaps the most important variable according to decades of research: social influence or the degree to which an individual perceives that important others want them to use an innovation, such as family, friends, colleagues, a celebrity that they have a parasocial relationship with, advertising. But sometimes you feel that your environment won’t facilitate your adoption. Facilitating conditions are the degree to which an individual believes that an organization and technical infrastructure exists to support the use of the system. The question to answer here is: How much will it cost? What happens if it breaks down? Why should I trust the developers, et cetera? Finally, the strength of these variables’ influence is moderated by age, gender, experience, and voluntariness. Men and women, as a broad example, interact with technology differently. They perceive risks differently, often have different priorities. Age is also a massive moderating factor. Younger folks are often more open to experimenting with new computer systems than older folks. But it’s often older folks that have the economic clout to decide what innovations are adopted into organizations. There may exist less perceived performance enhancement among older people, who are already successful in ag or have a higher expectation as to the effort they’ll be required to apply if they’re trying to learn a new computer system. And then there’s voluntariness. The orchard worker has no choice but to use an automated system if their boss decides to buy and incorporate one into their practice. So unfortunately, it doesn’t really matter what they think according to this model. You have to use the system or you have to find a new job. Experience with the system is a factor. Performance expectancy and effort, et cetera, can all be enhanced through exposure and hands-on experience with the system. Try before you buy. Talking about AI with someone with no conscious experience using AI is a lot different than talking to someone with lots of experience. Know your audience and establish experience level. All of this might give you some big clues as to what you want to be discussing or considering yourself in regard to any new idea or innovation in ag. Your audience, whoever they are, whatever their relationship with technology, needs information on performance. Yes, that of the system, but also a sense of how they might perform personally if they use the system. They need information on effort expectancy and to know, realistically, the learning curve involved with deploying a new system. They’ll want to know the extent to which it’s been used by people like them and what people like themselves say about a system. And they’ll want to know what structures and conditions are in place, what external factors exist that can enable or prevent them from appropriately deploying and maintaining a system. This is a good starting point when strategically planning how and what to communicate with your audiences about AI or any other new system or innovation. | Text on screen “tech adoption on the individual scale.” To the right of the screen two green boxes both with white writing. Box one reads, “usage intention.” An arrow points from this to the next, which reads, “Behavior.” To the left of the screen and above the other boxes, a blue box. Inside it text reads, “Performance expectancy.” Below the blue boxes, a third blue box. Text inside reads, “social influence.” A 4th blue box reads, “facilitating conditions.” All four boxes have arrows pointing to the other boxes. An ice blue box appears below and center. This box reads, “age, gender, expectations and Voluntariness.” It’s four arrows point up to the other arrows. In the studio, Alex. |
| [ Music ] Here are the key insights to take from this video: We can use diffusion of innovations theory and our audience knowledge to tailor our approach for communicating with different adopter groups. We can determine who is adopting and who is likely to adopt next by establishing what percentage of any given population has already adopted a particular innovation. By learning about your target audience, you can develop communications that incorporate the type of information they need and the limitations of discussing adoption at a particular point along the adoption curve. Research also tells us what kind of factors potential adopters of all types are likely considering in relation to new computer technologies, like AI, specifically. They consider how it might enhance or diminish their personal performance. They consider how much effort it will take to learn and use a novel system. They consider who else is using and what socially influential people and organizations have to say about the innovation. And they have to consider external factors, like cost, repairs, technical support, et cetera. You can influence all of these variables through purposeful communication and the type of information you provide for your audience. | Text on screen. “The takeaways.” An electric light bulb burns bright. Inside it stems grow up and out of the glass to become green leaves. [ Music ] Beside the bulb, bullet points appear. “Different adopter groups have different information preferences.” “Innovators are receptive to technical communications.” “Early adopters seek proof of concept before taking a risk.” “Early majority want to see reliable wins.” “Late majority need strong assurances.” “Laggards may never adopt or do so only out of necessity.” More bullet points read, “Adoption decisions are informed by key perceptions.” “Performance expectancy.” “Effort expectancy.” “Social influence.” “Facilitating conditions.” |
| [ Music ] Social science and empirical communication research are valuable tools for any professional communicator to explore. These and other models can really help hone your communication strategies and ensure you address the factors that are likely most influential to your audience’s decision-making, thus most useful to them. I hope this video inspires you to explore more psychological models and add to your communication arsenal. Thank you for your energy and attention. | In white over black text reads, “The closer.” In the studio, Alex. |
| [ Music ] | Text on screen appears over archival footage of lush green fields. “Presented in the Public Interest by United States Department of Agriculture.” The logos for “National Institute of Food and Agriculture U S DEPARTMENT OF AGRICULTURE,” “Western SARE,” and “Ag Aid Institute.” Text on screen reads, “This material is based upon work that is supported by the National Institute of Food and Agriculture, U.S. Department of Agriculture, under award number 2023-38640-39571 through the Western Sustainable Agriculture Research and Education program under project number WPDP 24-013. U.S.D.A. is an equal opportunity employer and service provider. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture.” More text reads, “This material is based upon work supported by the AI Research Institutes program supported by NSF and USDA-NIFA under the AI Institute, Agricultural AI for Transforming Workforce and Decision Support, (Ag AID). Award Number, 2021-67021-35344.” |