Even as technology and artificial intelligence (AI) transform the pork industry, Dan Hamilton, senior director for product performance in the Americas for PIC, argues that a curious human mind is more important than ever.
“As an industry, we must be open-minded, especially toward new technologies and new ways of doing things,” Hamilton says. “To remain competitive, we must have the curiosity and willingness to try things so we can stay on the cutting edge and be right where we want to be.”
At the forefront of significant advancements in swine genetics, Hamilton has witnessed how cameras, sensors, and AI are revolutionizing the barn. While these tools collect and process massive amounts of data at lightning speed, Hamilton warns against blind reliance on the “black box” of technology.
“I always tell people when you create an algorithm, it always gives you an answer,” Hamilton says. “But stay curious. Ask yourself if it’s the right answer and if we should be making a decision based on it.”
Validating the Algorithm
Nothing replaces an inquisitive mind that constantly asks how to make things better. For Hamilton, the power of a company like PIC lies in its access to large, commercially relevant data sets. While PIC has always relied on nucleus farms for precise data, Hamilton notes they have expanded their reach into the commercial sector.
“As we use these new technologies, we’ve got more data points in differing environments,” Hamilton explains. “We must continually validate these technologies to ensure they are bringing value, and we must continue to retool the algorithms to make them more accurate and refined for their specific locations.”
Taking the Subjectivity Out of Selection
Digital phenotyping—the use of automated technologies like cameras and sensors—allows producers to measure behavioral, structural, and production traits objectively and non-invasively. Hamilton has focused specifically on how this technology evaluates the feet and legs of swine.
“Feet and leg quality are vital to a sound breeding program,” he notes. “For generations, good stockmen have selected for better feet and legs, but there has always been a level of subjectivity. What I think is ideal, you might see as slightly different.”
By training AI algorithms using images from expert selectors, researchers can now evaluate structural soundness in real-time.
“This is a huge step forward,” Hamilton says. “We’ve found that algorithms are often more accurate and consistent than humans. Even the best selector doesn’t always score the same animal the same way every time. It could be because it’s Monday morning versus Wednesday afternoon, or they caught the pig at the wrong moment. The cameras, however, are three times more accurate, allowing us to make faster genetic progress.”
The ultimate goal? Predicting longevity.
“We are now using that data to predict which gilts or boars will have the greatest longevity in the sow herd,” he says. “Moving from what we thought were the best feet and legs to actual data based on herd retention is a major shift toward better outcomes.”
Decoding the “Social Network” of Pigs
The modern pig is not the animal your grandparents raised. Today’s pigs are raised in large, indoor pens, meaning selection traits must evolve alongside the environment. While profitability remains a driver, Hamilton says welfare and behavior are becoming equally critical.
“When I was in graduate school, we used to sit pen-side and manually record pig behavior on a tablet,” he recalls. “We could never get large enough data sets to really move the needle. Today, with AI, we can obtain behavior recordings on thousands of pigs simultaneously.”
By measuring “normal” behaviors—time spent eating, lying, sitting, or drinking—PIC is identifying highly heritable traits that can be improved through selection. This has led to the study of “social networking” within the pen.
“Pigs develop behaviors for a reason,” Hamilton says. “Understanding how they live together—which ones are dominant, which are subordinate, and what ‘cliques’ they form—creates opportunities. We want higher-performing, more profitable pigs, but we also want pigs with behaviors that are better for their pen-mates and their caregivers.”
Precision at the Feeder
Beyond health alerts and activity monitoring, Hamilton sees a future where cameras count pigs and predict weights with near-perfect accuracy, ensuring pigs go to market at the optimal size according to their growth curve.
He is particularly excited about the potential for AI to improve sow livability through automated body condition scoring.
“If you can take caliper or body condition readings and have them ‘talk’ to an automatic feeding system, you manage daily intake based on the sow’s specific condition and weight,” Hamilton says. “That brings us to a level of precision we’ve never seen on the farm.”
The Challenge of the System
Despite the high-tech tools, Hamilton reminds producers that they are still managing biological beings within a complex production system. Challenges like health breaks and “flow constraints” remain the industry’s biggest hurdles.
“Producers often have to put X number of pigs into X number of spaces. If they have a few extra pigs, they still go into that same space,” he says. “What is profitable in the short term may not allow for maximum genetic performance.”
Hamilton also applies his “curious mind” philosophy to the ongoing debate over pork quality. He suggests the industry needs to stop looking only at the loin and start looking at the consumer.
“We have great tools to improve pH, color, and marbling in the loin, but the loin is only one part of the carcass,” he says. “If we improve loin quality at the expense of ham quality, are we really winning? We need to understand why the U.S. consumer isn’t paying for higher quality before we can truly decide how fast to move. The technology is here, but the questions, and the curiosity, must come first.”
You can listen to more of Hamilton’s personal story of resilience on “The PORK Podcast” on YouTube or follow The PORK Podcast anywhere podcasts are found.
