When most people talk about food waste, they’re usually referring to the food we throw into the garbage or food waste bin, whether that’s a wilting head of lettuce or a bruised banana.
But what about the extra “invisible” food sent to lower-value, non-human-consumption destinations such as rendering or composting? This type of “waste” occurs on a large scale in processing plants around the world, where tiny inconsistencies in how operators carve meat translate into enormous volumes of edible protein being diverted to rendering. In other words, protein that could otherwise go toward someone’s dinner is instead directed into lower-value products like tallow or pet food ingredients.
What if we could use technology such as AI to put more food on our tables? That’s exactly what processing giant Cargill is doing with its new platform called CarVe. CarVe is an AI-powered system designed to eliminate these invisible losses before they happen. The system just won a Bronze Edison Award and is one of the more compelling examples of AI delivering measurable food waste reduction at industrial scale.
“We have these protein facilities where operators are carving pieces of meat in real time,” said Abhishek Roy, who leads AI initiatives at Cargill. “If you leave meat on a ribeye, that translates to a lot of loss, because all of that goes to rendering. That’s also food waste, because that could go onto someone’s table.”
As of April 2026, Cargill has deployed CarVe across three primary beef sites (Friona, Texas; Fort Morgan, Colorado; and High River, Canada), with four more planned. CarVe is helping operators as they process up to 3,500 cattle moving through the line each shift. According to Roy, the system is already delivering value capture in the millions of dollars and recovering thousands of pounds of meat that otherwise would have gone to rendering.
Here’s how it works: CarVe utilizes computer vision by placing cameras along the processing line and providing operators with real-time, color-coded feedback on their cuts. The system assigns each operator a carving score, and as workers improve, the yield efficiency gain is typically 3 to 5 percent per cut. According to Cargill, even a one percent yield improvement across that volume could save hundreds of millions of pounds of meat.
CarVe was just one of many initiatives I wrote about in a new report on AI’s impact on food waste, which I authored in partnership with ReFED. For the report, I interviewed more than 35 experts, including AI product builders, operators, and industry leaders. Before this report, there really hadn’t been one that examined AI’s impact on waste across every part of the food value chain (I know because I looked), and it is arguably the first comprehensive, system-wide report on AI and food waste.
What I found in writing the report is that food companies big and small are increasingly using AI to make their businesses more efficient and less wasteful. In foodservice, companies like Leanpath and Winnow have documented 20 to 53 percent waste reductions using AI-powered tracking systems. In retail, Afresh has reportedly helped prevent more than 200 million pounds of food loss. Upstream, companies like Strella are reducing waste by combining AI with IoT sensors, while others like Driscoll’s and Simplot are using AI to accelerate the development of climate-smart crops with technologies like gene editing to ultimately design some of the waste out of food.
If you’re interested in the full report, it’s available on ReFED’s website.