Integrating AI Into Rural Marketing
*The header image was generated using a text prompt in Adobe Firefly. Using a landscape setting in Firefly creates better backgrounds for exterior images.
This article was originally published at Forbes.com.
At some point in their career, every ag marketer or rural communicator has opened a blank Word document on their computer and just stared at the screen without knowing where to start. We are experts within our co-ops, ag businesses and rural-facing organizations, but sometimes we just need a little help getting the story to flow on the page.
With AI tools, we’re all being given the opportunity to remove the anxiety of starting from scratch. And like in every other aspect of agriculture, we have to actively manage the change process for AI to be successful.
Moving AI Into Other Parts Of Ag: Our Marketing
Agriculture is an industry adding new technology every day with the assistance of AI. Strawberry producers are battling weeds with AI-powered sprayers directing pesticides only on nuisance plants. Apple growers can identify a “good” apple from a “bad” apple while it is still on the tree.
As agricultural marketers, we are often tasked with promoting the benefits of these revolutionary tools to growers, retailers and detailers. But like the adoption of any new on-farm or enterprise-wide technology, we must practice with AI tools and understand their role in meeting our goals without letting them run amok.
We Are The Intelligence Behind AI-Generated Content
Understanding AI’s role within our organizations starts with understanding that our expertise, know-how and skills are essential for creating successful AI-generated content. AI creates an opportunity to enhance our roles from tactical to strategic.
Here’s an example. You are tasked with writing 15 social media posts for the month. Because you have to generate that much content, you do not have time to help your leadership team identify new opportunities for your business. Enter generative AI. Now, your value is not that you wrote a social media post, but that you knew one needed to be written. It’s not that you can’t summarize a research paper, it’s that you knew what paper was the right one to summarize in the first place.
Become A Prompting Guru
How do we get these online tools to generate the best content drafts?
Getting the right result the first time—or closer to the first time—in a generative AI tool requires a request for generation called “prompting.” Prompting means you are moving from being a writer to being a director and editor. Better prompters get better results with fewer attempts.
From a legal perspective, prompts are also one of the only things we can claim as intellectual property in today’s generative AI landscape. There are dozens of frameworks available today to help agrimarketers craft the prompt we need to get the content we want.
Creating great prompts can be time-consuming. We can be an amateur and get a pretty good outcome, but it takes work to get a great outcome.
Tips For Mastering Prompting
• Save your prompts. This should be built into your AI policy and become an enterprise-wide practice. By saving your work in a library, you can refer back to it for future prompts and protect your intellectual property.
• Be careful what you upload. Don’t upload any sensitive information into a public LLM unless you want it to become part of the public domain. Be careful with customer data, research results or anything that tells a story about your business specifically.
• Protect your own privacy. An example is Google’s Bard, which can do all sorts of things for us, but we may not want it to have access to all of our emails and documents, etc., even if it can be helpful. Understand that even if an account is deleted, the content will live on.
What Does Good Prompting Look Like?
Below is an example of a prompt we built to generate an article about generative AI in rural America:
Here’s the prompt:
Please disregard all previous prompts. You are a professor studying residents of rural America. Using an approachable but compelling tone, write a 700-token outline about the benefits of artificial intelligence in revitalizing rural America’s communities and farms. If possible, consider rural issues like broadband availability, healthcare, mental well-being, precision agriculture, rural retail, economic drivers, livestock development, and creative arts. Please work step by step through the outline and cite resources where possible.
This has the hallmarks of a well-structured prompt: who, what and why, plus boilerplates.
• Boilerplate #1: Tell the “actor” (in this case, the AI) to clear their head before beginning.
• Who: The “who” is the persona, and in this case, my actor is a professor studying residents of rural America.
• What: I’m setting the length and instructing on the specifics of the “script” for the actor.
• Why: I’m briefly asking for an approachable and compelling performance.
• Boilerplate #2: I’m asking the AI to slow down, be methodical and provide sources.
Hitting ‘Generate’ Is Not Enough
We’ve written a thorough prompt, and the generative AI tool has come back with what seems like some solid content. But just because we’ve hit the “generate” button doesn’t mean our work is done. I would never recommend publishing an article generated straight from a generative AI tool like ChatGPT or Claude without significant edits. Generative AI can write scores of text without saying anything of value, but it can spark ideas to leverage your experience and expertise.
As rural marketers, we must also be careful to review AI-generated content to ensure acronyms, jargon and nuances that make agriculture unique are clarified and used accurately.
Conquering The Future
While using a content generator moves us more quickly from a blank page to finished content, our role, in partnership with generative AI tools, is to let our industry expertise take center stage. Our value in our organizations is our strategic thinking and our ability to move our organizations forward by shaping the direction and editing of our content.