About a month ago, my husband and I watched a fascinating — albeit slightly terrifying — segment on 60 Minutes about artificial intelligence (A.I.).
One A.I. program the episode focused on was IBM’s Watson, which 60 Minutes deemed “one of the most sophisticated computing systems ever built.”
It’s no surprise why. According to the show, Watson is able to consume the equivalent of a million books per second. After winning a game of “Jeopardy!” in 2011, IBM decided to test the system further by enlisting 20 top cancer institutes to tutor Watson in genomics and oncology.
Today, the system is helping the doctors at the cancer center at University of North Carolina at Chapel Hill come up with treatment options for cancer patients who already failed standard therapies.
Dr. Ned Sharpless, who runs the cancer center, has been impressed with some of the treatment options Watson has come up with that the medical team hadn’t already thought of, simply because it’s able to keep up with the thousands of new research papers and trials that are released daily.
With more no-tillers and agronomists relying on precision technology for their management and decision-making, this got me thinking — how could Watson and other A.I. systems influence agriculture?
A.I. has already made its debut in some parts of our industry. E&J Gallo Winery in Modesto, Calif., is testing a new irrigation system with IBM that will use less water, with the intention of taking what’s learned and teaching it to Watson to help growers around the world.
In Europe, Danish ag engineers have built a self-propelled robot that can recognize 25 different weeds and spray them directly — reducing herbicide usage by up to 75%.
If Watson can help doctors come up with new treatment options, we can assume it could help the EPA determine the safety of current and future pesticides.
As executive editor and publisher Darrell Bruggink mentions in his recent blog on visiting Syngenta’s headquarters, there are more than 7,000 studies just on atrazine. If the EPA could employ Watson to read and interpret all of the studies out there on a certain pesticide, registrations may be approved or denied faster, and the agency could have more confidence in its decision on a product.
The same goes for genetics. What kind of improvement could we see in our corn hybrids, soybean varieties and other seeds if seed breeders had a computer like Watson offering suggestions?
Not to mention at the farm level. What if you could teach Watson or another A.I. program to read a farm’s history, including yield maps, fertilizer applications, populations, etc.? Imagine the detailed prescriptions it could come up with.
In short, the possibilities with A.I. seem endless. While I’m not sure when we can expect this level of technology to make a greater appearance in agriculture — and I’ll admit the movie “2001: A Space Odyssey” still has me a little apprehensive about A.I. — the potential it can have on our industry and the world seems promising.