What Can Growers Do With AI Now?
February 28, 2026 Artificial intelligence (AI) is everywhere and in every conversation in some form today. It is impacting many industries.
In the cultural AI conversation, there are plenty of doom sellers and naysayers and sycophants and utopians and everything in between. Somewhere between the doom sellers and the utopians, real work is getting done with AI, and real, substantive improvements in work process efficiency and capability are being made.
Some tasks and industries are being completely transformed by the capabilities of the new AI tools coming online. Others are not yet impacted directly by the changes.
The question on the minds of many growers I talk to is, what can (or should) we do with AI now? Many growers just want an easy way to start to use the tools and begin to understand how they work and what they can do. One good first way to approach these tools is to use AI now like an information assistant.
Research and Search for Answers
AI Services like Claude, ChatGPT, Perplexity, or Gemini (Google) can speed up the search and synthesis for information you are looking for or trying to learn.
Traditional web search presents links that have the highest relevance score for your search terms (or the highest bidder). AI chat bot services not only search but also synthesize answers across the many links and sources on the internet they were trained with. This speeds up getting the information you need in an easily accessible format.
The key difference is that we are used to getting a Google list of places (web sites) to potentially get the answer to our question or search term versus getting a constructed answer by the AI bot.
I also use these services regularly to do deeper web research on technical and industry topics and news. I have observed that tools like Claude.ai, Perplexity, and ChatGPT do good job with scanning and summarizing up-to-date information. Some models, however, due to their training window (the end of the current calendar windows their training is limited by), aren’t updated with some of the very latest developments.
Content Generation
AI language model chat services also do well at generating content. If you need to write some sales copy or want to create a marketing email that hits on certain points, these tools do well at creating those drafts and refining them. I have used them to draft and iterate the content of business correspondence in cases where I was struggling with the right wording.
There are also many AI services that can also create images, audio, or video clips with a prompt.
I have personally found ChatGPT to generate better imagery for my needs and I usually let it have a go at any content generation like letters and such. Anthropic does a good job with text, but I find its image generation to be less than what I need.
Technical Work Assistant
I use AI language model tools like ChatGPT or Perplexity as a technology assistant to dig into and evaluate (summarize) very specific technical research and provide information, papers, and training on specialized technical topics for my own work, our team, and our products.
Further, the coding-related AI tools allow me to draft and test program code, validate math and statistics assumptions and procedures, and validate or perform in-depth analysis (assuming access to the constituent and dependent data).
In these cases, when tasked with a deeply technical task, the AI bot will usually create program code in a coding language, which it will then attempt to execute to generate the specific view or answer. For teams like ours, this is natural. For those who are not experienced with software program coding, this can be surprising since they are not familiar with the syntax of programming languages.
Specialized Third-Party Tools
Beyond the basic chatbots, there are a variety of other tools that embed AI to improve key workflows. There are now many options that incorporate AI capability for speeding tasks, reducing time, or improving efficiency.
Our team has used or experimented with various tools to improve the efficiency of our departments. Here are just a few examples of AI-enabled tools that we have tried and found to be helpful for some of our company’s specific use cases.
- Customer Support – Intercom or Freshdesk
- Visual Design – Figma Make or Canva
- Sales Deal management – Attio or Hubspot
- Credit Card Management – Ramp
- Meeting Transcription and Summary – Fireflies or Granola
Clearly, this is barely scratching the surface on what is now becoming available to help grower needs in some of the general work tasks.
Other AI Considerations
AI is still wrong sometimes. AI is certainly a powerful tool, and modern language models like ChatGPT, Gemini, or Claude can do impressive things. But they still make mistakes.
Recently, I was doing a comprehensive pricing analysis for a large U.S. grower and wanted my AI tool to check one aspect of my math in the analysis just to make sure I was right, as it was a very complicated formula. So, I asked the chat bot to do a check, and it spit out an answer that was clearly not right. The following image is from that conversation (I am “J” in the conversation, and the response is from the AI bot):
Image credit: John Beauford
After I pointed out that the AI chat bot was wrong, it was able to re-do the math and get the right answer. The lesson? Always verify the work of your AI chat bot.
Sometimes AI Can Ce Funny
A few weeks ago, I was using a coding chat bot to verify some software changes for one of our analytical products. As I was reviewing the work, it seemed to me that the AI bot had removed some code that it was not supposed to, since I could not see the evidence it worked. Here is the actual conversation with the chat bot.
Image credit: John Beauford
Yes, I missed that in my own code. Yes, I was a little cheeky, mostly just to see what it would do. Yes, this is a literal cut and paste. And, yes, the AI language model really said, “We are all human”. Let that sink in. 😊
As you use these tools, sometimes the textual results can be funny.
A Note on AI Security
In my conversations with growers around the country, a consistent theme about AI is security. I get queries from growers like:
- “Is my data safe?”
- “Will my company data be used to train AI?”
- “Will my competitors be able to see or use my data?”
- “What does AI do with my data?”
And many more.
The fundamental issue is “Is my data safe with AI?”
I am afraid there is not a single, precise answer to that question, it truly depends.
The specific use of data or text you load into or type into a commercial AI chatbot depends on the terms and conditions of the AI chat bot and the company that owns it. The legal terms can vary widely between companies, tools, usage tiers, or plans.
In general, most AI chat bots can use user generated data and text to update model training. This is especially true for the free tiers of usage of many popular chat bots. As you upgrade to higher subscription tiers or migrate to some of the enterprise usage levels, some of those terms and conditions are modified to exclude data from training or retention.
The best advice I have is to review the specific terms and conditions of the AI chat bot or tool you choose to use and the level you subscribe to, and understand explicitly what the terms and conditions will allow. (AI can help with this!) Then, once you and your team know that, make sure your employees’ usage (what you tell and load to AI) is in line with that.
Each of the major chat bot companies will have their terms and conditions somewhere on their website or it will be made available when you sign up for a plan level.
So, check the terms and make sure you and your team are comfortable with what is done (or not done) with your prompts and data.
AI Now
Between the doom sellers and the utopians, AI capabilities are making practical and measurable improvements to information work at a very fast pace. Growers and their staff can benefit from incorporating these new technologies and tools in their workflows.
The AI landscape and capabilities are growing and changing rapidly. With the right tool or plan, in line with a grower’s privacy needs, they can be a big efficiency boost to information heavy workflows.
A Note on Web Searching
I just want to make a call out to remember the writers. Claude, ChatGPT, Perplexity, and the other AI engines would be nothing were it not for the millions of men and women who wrote books and papers and blogs and articles and made them available. That information is the fodder of AI training. So don’t let AI convenience block clicking through and visiting web sites and reading the original sources. As someone who has written a lot of web content, I and all my fellow authors would appreciate it.
Subscribe to eNewsletter
