FROM FARM TO FORK:
HOW AI IS TRANSFORMING FOOD SAFETY

Learn how AI is being used in farming to eliminate pesticides in the U.S. food supply Learn more

BACKGROUND:
Fresh fruits and vegetables are essential to a healthy diet. However, much of the produce grown in the United States has tested positive for pesticides, even after washing. Farmers use herbicides and pesticides on crops to protect them from weed and insect infestation, but chemicals are now in the food supply. The Environmental Working Group’s (EWG) 2026 Shopper’s Guide to Pesticides in Produce compiled data from pesticide residue testing conducted by the United States Department of Agriculture (USDA), covering more than 50,000 samples of 47 types of fruits and vegetables. It found that 96% of the “Dirty Dozen” fruits and vegetables tested positive for pesticides, with spinach topping the list. Others included strawberries, apples and potatoes.

So, what can be done to limit the consumption of these chemicals? AI now offers farmers a way to eliminate herbicides and pesticides entirely.

• Carbon Robotics is revolutionizing agriculture with AI and robotics to reduce costs and increase yields. Amid labor challenges, rising input costs, and increasing concerns about herbicide use, growers worldwide are seeking smarter, more efficient ways to farm. Carbon Robotics addresses these needs by delivering advanced AI robotic laser weeders and tractor autonomy that drives efficiencies to the global community. Leveraging the full-stack of NVIDIA hardware and software, Carbon Robotics’ LaserWeeding technology targets the weed's growth center with a laser, disrupting cellular growth and preventing regrowth. Each of our LaserWeeder machines has 24 NVIDIA GPUs onboard, allowing it to analyze thousands of plant images per second in real time. We currently use the NVIDIA RTX 4000, with an upgrade to the more powerful RTX Pro 4000 planned for later this year. Our autonomous tractor attachment, the Carbon ATK, uses a different NVIDIA chip called the Jetson Orin — a compact, energy-efficient processor purpose-built for robots and self-driving machines.
• We use two key pieces of NVIDIA software that make our AI run faster. The first, CUDA, lets our software tap into the full power of NVIDIA's processors — think of it as the engine that lets us crunch enormous amounts of data quickly. The second, cuDNN, is a specialized add-on built on top of CUDA that's specifically optimized for AI and machine learning tasks, helping our models identify and target weeds with greater speed and accuracy.

The benefits are significant. This technology does not disrupt the soil like mechanical weeding can do. For consumers, this results in healthier fresh vegetables and herbs that are not sprayed with chemical herbicides, for both organic and conventionally farmed produce. For the environment, this eliminates chemical herbicides from the soil and prevents runoff of these chemicals into our rivers and lakes. For farmers, this results in lower costs and higher crop yields, quality and consistency.

In this segment, Carbon Robotics founder and CEO Paul Mikesell explains how AI is revolutionizing the farming industry and keeping our food free of pesticides.

For more information, please visit: https://carbonrobotics.com/

MORE ABOUT PAUL MIKESELL:
Paul Mikesell is the founder and CEO of Carbon Robotics in Seattle, Washington. Prior to Carbon, Paul was responsible for scaling Uber’s backend systems, opening the Seattle engineering office, and later focusing on Deep Learning and Autonomy. Paul also co-founded Isilon Systems, a distributed storage company, in 2001. Isilon went public in 2006 and was acquired by EMC for $2.5 billion in 2010. Paul holds a bachelor’s degree in computer science from the University of Washington.

Produced for: Carbon Robotics & NVIDIA