ChatGPT for Precision Ag? Not Yet.

This is Tim Hammerich of the Ag Information Network with your Farm of the Future Report.

Almost every industry is using some type of artificial intelligence capability these days, so where can this be useful in precision agriculture? Dr. Steve Shirtliffe, professor at the University of Saskatchewan, says they’ve learned with plant and soil science, AI models are only as good as the information you feed it.

Shirtliffe... "Even though there are things like the Google Earth engine, the good data sources are often fragmented and you need a lot of what we call domain knowledge to find and interpret them. People like myself and Preston and the people that work in my lab to be able to understand what data sets you should apply to what problems. So I think we're a ways off from really having a kind of a ChatGPT precision ag where you could just ask how much nitrogen should I put on my field this year, you know. That's a long ways away, but using machine learning to make sense of that data when we find it is just standard procedure."

But one area AI really comes in handy is analyzing large data sets adds Dr. Preston Sorenson, professor of soil science.

Sorenson... "So because our data sets are large and then there's always inherent noise in the data sets, we have to be careful about the types of machine learning models we use. So I would say for now, some of the really big advancements in AI generally, we're still not able to take advantage of yet, at least not until we get larger, more comprehensive soil data sets."

Both professors say at least for the foreseeable future, boots on the ground data collection is still needed.