The big buzzword of 2025 is “AI” or artificial intelligence. Everyone is talking about how to use it, where it fits, and whether we should be using it at all. For dealerships, AI could be genuinely helpful, but only if we use it with real understanding instead of treating it like a trend. That means deciding when you’ll use it, how you’ll use it, and setting clear guidelines, training, and expectations. There are real risks, limits, and strengths you must understand.
Safety First, AI Out of the Box
In a recent online seminar, someone asked me what I thought about agriculture using AI and large language models. I told them that you should treat AI the same way you treat social media. These tools learn from the information you type into them, which means anything you share can eventually become part of the system and passed along to others.
Additionally, unlike a doctor, lawyer, or a secure shared drive, AI doesn’t have legal protections to keep your information private. What you enter can be accessed later in ways you might not expect.
Even when the AI isn’t quoting anyone directly, its answers are shaped by millions of decisions, opinions, and pieces of data. Just like you wouldn’t want weather data from New Mexico when you farm in North Dakota, AI can also give wrong or misleading results because it fills in gaps with guesses.
So, as your dealership starts using or promoting AI, keep this in mind. AI can be a powerful tool but only if you’re transparent about when you use it and how you use it.
It’s a prudent idea to create an AI strategy and policy for your dealership. Think about:
- How AI will be used
- How grower data will be protected
- What you’ll do if AI gives an answer that leads a grower in the wrong direction
These decisions need to be made before AI becomes part of everyday work.
How Do You Use AI & Are You Automating the Right Things?
It’s easy to get excited about automating everything. But there are tasks that shouldn’t be automated and that’s where the concept of “planned friction” matters. Planned friction is when you intentionally slow down a process, so people take time to think or evaluate the results. These are the areas where AI shouldn’t take over because people tend to pay less attention when a tool does the thinking for them.
Studies even show that people remember less information when they use AI to help them than when they do things the traditional way.
Choosing where AI fits is just as important as choosing the tool itself. AI also comes with costs beyond money, like environmental impact, privacy concerns, and the risk of losing critical thinking skills. AI can automate information, but it can’t replace wisdom.
Train Your Staff to Use AI Well
Effective use of AI depends on good training. Staff need to know what counts as proper and improper use, and they also need to learn how to “prompt” AI in a meaningful way.
People who treat AI like a Google search box usually get weaker results. Training should cover:
- How to ask better, clearer questions.
- How different prompt styles change the results you might get and the formatting of information.
- How to refine questions or provide data to improve results that AI will provide to you.
- How to explain how the AI came to its conclusion by authenticating its sourcing.
- How to show growers a confidence score or reasoning summary.
With the right training, AI becomes more useful and more dependable.
How AI Does or Doesn’t Work
Like any innovative technology, AI takes a concerted effort to practice. Most first-time users treat it like a search engine, but it’s not built for that. Beyond privacy issues, there’s also the risk of using AI to replace professionals.
For example, AI can analyze a spreadsheet, but it doesn’t have the judgment of a trained statistician. It can compare agronomic data to a curve, but it doesn’t know which statistical method is truly proper.
It also helps to learn different prompt design methods. Prompt engineering such as being specific, adding context, choosing tone, refining your request is a skill. Trying different prompts for the same task can help you understand AI much better.
Once you get comfortable, you can start figuring out the real ROI for using AI. Look for tasks that take too much time or are overly repetitive without any input value. But don’t forget not all “wasted time” is bad. Planned friction helps us slow down and make better decisions.
Decide which tasks can rely on automation (like filling out standard forms) and which ones need human attention (like reviewing data or making decisions).
Is There a Dealer Opportunity?
So, does AI have a place in your dealership? Maybe — but not automatically. Each dealership must look at both opportunities and risks.
Growers evaluate precision ag tools carefully, and dealerships should treat AI the same way. An 8-row planter is useless to an orchard grower, even if it’s full of the latest tech. AI might not be right for every dealership, or it could be that your teams just haven’t seen the value yet.
A fair, realistic, well-researched approach will help you figure out what works for you without getting swept up in the hype or shutting the door too quickly to innovations only to be left behind in inefficiencies. With the right mindset, AI really can help move your dealership forward.
