AI in agriculture: Experts say human judgment remains key as technology advances
Sept. 28, 2026
By John Lovett
University of Arkansas Division of Agriculture
Fast Facts
- AI increasingly being adopted in systems already in use on the farm
- Limitations of AI judgment expected to continue
- Need for students to gain access to technologies they will likely encounter in industry
(913 words)
Download PHOTOS of roundtable panelists
FAYETTEVILLE, Ark. — As artificial intelligence becomes faster and more capable of complex tasks, human critical thinking and the ability to ask the right questions will become even more important.
At the AI in Agriculture Symposium on Sept. 21, panelists in a roundtable discussion noted that many tasks once considered specialized skills — like coding, data analysis and website development — can now be performed more quickly with the help of AI.
The symposium was hosted by the Center for Agricultural Data Analytics, a part of the University of Arkansas Division of Agriculture’s research arm, the Arkansas Agricultural Experiment Station.
On the farm, AI is increasingly being adopted in systems that are already in use, said Jason Davis, an assistant professor in the department of crop, soil, and environmental sciences and a remote sensing and pesticide application extension specialist for UADA.
“From my perspective, it has been primarily in row crops, embedded in systems they are already familiar with,” Davis said. “It’s a dashboard they’re already using, or a computer monitor in the cab of their tractor. It’s suddenly expanded utility and capability.”
For researchers who develop those AI systems, the panelists agreed that understanding the underlying science, recognizing flawed results and determining which problems are worth solving remain uniquely human responsibilities.
Several participants compared today's AI revolution to previous technological shifts that increased the value of higher-order thinking skills.
“AI is amazing. It’s a tool that is like a car. It makes you move faster, but we still need to know what is rational and understand what AI is doing,” said Arthur F.A. Fernandes, a geneticist for Cobb-Vantress. “We may not be coding as much anymore but we need to know how to read code and assess if that code is doing the right job.”
Tim Beissinger, co-founder and chief technology officer for Heritable Agriculture, related it to long division.
"When I was in elementary school I learned long division,” Beissinger said. “My mom was a math education professor, and she thought it was crazy that some schools weren’t even teaching long division anymore … but you really don’t need it. When I was a professor, I taught how to code in R, and maybe that’s like long division now, but you need to understand long division. You need to understand how to ask a scientific question.”
For students preparing to enter the workforce, the panelists encouraged a focus on scientific reasoning, problem solving and learning how to effectively evaluate AI-generated outputs rather than relying on the technology unquestioningly.
Looking ahead
Even as AI systems become more sophisticated, panelists said one challenge will remain constant: determining when results make sense and when they require further scrutiny. The ability to recognize errors, identify meaningful opportunities and understand stakeholder needs will also continue to distinguish human expertise from machine-generated outputs.
James Schnable, a professor and Nebraska Corn Checkoff Presidential Chair for the University of Nebraska–Lincoln, said he has not yet seen evidence of improvement on AI’s ability to find the right questions to ask or determine when a result looks right.
“It’s figuring out what is the thing people actually need,” Schnable said. “I think in the AI world, this is called taste as opposed to ability. And I have seen, if anything, the taste of AI — what questions it asks if you give it freedom. They’re less aligned with what we actually need. It’s actually getting worse.”
Dongyi Wang, an assistant professor of biological and agricultural engineering with UADA, spoke about the challenges of processing the massive amounts of information generated by AI.
“Our graduate students can produce a review paper in less than two weeks, but as a human it takes me a month to read hundreds of pages and carefully provide feedback because that’s still very important,” Wang said. “I would predict in the next five years, everyone will be buried in all kinds of information, not just text, but images and videos.”
The question Wang posed is: “How do we utilize that knowledge to distinguish what is needed?”
Some of that knowledge and data, however, may even be hidden in sources that AI can’t access due to formatting or being privately held.
“In our time, we have a responsibility to start this discussion of ‘how does everybody have an equal transparency in terms of data, not only on Facebook, but also with research data,” said Alexander Bucksch, an associate professor of plant science at the University of Arizona. “Why can you hold back 20 years of data that could already revolutionize the world?”
Preparing students for an AI-enabled world
Another recurring topic was how universities should prepare students for careers increasingly shaped by AI.
Rather than focusing solely on teaching programming syntax, panelists discussed placing greater emphasis on software design, problem formulation and the ability to work effectively alongside AI tools.
Speakers also highlighted the need for students to gain access to the technologies they are likely to encounter in industry, while maintaining standards for academic integrity and original thinking.
As the conversation concluded, participants were asked to describe AI in a single word. Their answers reflected the balance of optimism and caution that characterized the discussion: "trust," "amazing," "fun," "transformative" and "scale."
The roundtable was moderated by Leonardo Bastos, an assistant professor of integrative precision agriculture at the University of Georgia and co-host of The Crop Science Podcast Show, with Samuel B. Fernandes, an assistant professor of agricultural statistics and quantitative genetics with the Arkansas Agricultural Experiment Station and organizer of the symposium.
To learn more about ag and food research in Arkansas, visit aaes.uada.edu. Follow the Arkansas Agricultural Experiment Station on LinkedIn and sign up for our monthly newsletter, the Arkansas Agricultural Research Report. To learn more about the Division of Agriculture, visit uada.edu. To learn about extension programs in Arkansas, contact your local Cooperative Extension Service agent or visit uaex.uada.edu.
About the Division of Agriculture
The University of Arkansas Division of Agriculture’s mission is to strengthen agriculture, communities, and families by connecting trusted research to the adoption of best practices. Through the Agricultural Experiment Station and the Cooperative Extension Service, the Division of Agriculture conducts research and extension work within the nation’s historic land grant education system.
The Division of Agriculture is one of 22 entities within the University of Arkansas System. It has offices in all 75 counties in Arkansas and faculty on three system campuses.
Pursuant to 7 CFR § 15.3, the University of Arkansas Division of Agriculture offers all its Extension and Research programs and services (including employment) without regard to race, color, sex, national origin, religion, age, disability, marital or veteran status, genetic information, sexual preference, pregnancy or any other legally protected status, and is an equal opportunity institution.
# # #
Media Contact:
John Lovett
U of A Division of Agriculture
Arkansas Agricultural Experiment Station
(479) 763-5929
jlovett@uada.edu