María Pilar Bernal Research Professor, Centro de Edafología y Biología Aplicada del Segura, Consejo Superior de Investigaciones Científicas Murcia Kristie Ebi Professor, Environment and Occupational Health Sciences, University of Washington J P Tripathi Director, Agriculture Programs, Wadhwani Institute for Artificial Intelligence In 2022 alone, households generated more than one billion tonnes of food waste, much of it ending up in landfills or other unsustainable disposal routes.20 Recent progress in automation and robotics – driven by advances in AI, machine learning and computer vision – is making it easier to separate food from other waste streams, enabling large-scale recovery for composting, biogas production or upcycling.21 By scaling up diversion and reuse, food waste automation technologies reduce landfill emissions, lower demand for newly sourced agricultural inputs and strengthen circular food systems. This helps to protect planetary boundaries related to climate change, land-system change, biogeochemical flows and biosphere integrity. Recent improvements in image recognition and waste classification algorithms enable automated sorting systems to identify food waste even under challenging, real-world conditions – including spoiled, processed or partially hidden items.22 These systems often combine advanced imaging techniques (e.g. near-infrared and hyperspectral imaging) with robotic arms, with companies like ZenRobotics using real-time data to separate food from packaging or other contaminants with minimal human input23 – producing cleaner waste streams suitable for composting, anaerobic digestion or upcycling. Once limited to large recycling plants and agricultural processors, advances in sensor design and energy efficiency have made the technology more compact and affordable, allowing broader deployment in municipal waste facilities. In Seoul, automation supports a citywide food waste programme that diverts over 95% of food scraps from landfill or incineration.24 The Netherlands are testing AI-driven sorting lines to separate food from packaging at centralized composting hubs.25 The US and Europe are piloting modular units in grocery stores, university campuses and commercial kitchens to recover waste at the point of generation with companies like Orbisk.26,27 Growing demand for upcycled products – including animal feed, compost blends, bioplastics and even food ingredients – is increasing the economic viability of these systems. A recent estimate valued the global market for upcycled food products at over $50 billion, with strong growth expected over the next decade.28 Automated sorting systems may transform waste management industries, with wide-ranging environmental, economic and societal impact. Environmentally, these systems reduce methane emissions from landfills and support nutrient recycling through composting and anaerobic digestion, helping to protect planetary boundaries related to climate change and biogeochemical flows. Economically, automation could open new markets for upcycled food waste products such as animal feed, compost blends and bioplastics – while creating new roles in sensor maintenance, robotics integration and circular product development. However, high capital costs may limit adoption in low-income regions or small-scale facilities, slowing widespread impact. Societally, expanded recovery infrastructure could reduce environmental burdens in low-income communities disproportionately affected by landfills, while strengthening city-level efforts to meet zero-waste and climate targets. In doing so, food waste automation could help close a critical gap between growing waste volumes and limited capacity for sustainable disposal.95% of food scraps in Seoul are diverted from landfill or incineration through a citywide food waste programme powered by automation. 10 Emerging Technology Solutions for Planetary Health 14

10 Emerging Technology Solutions for Planetary Health 2025