This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility: Artificial intelligence promises to transform food safety by helping companies identify risks earlier, predict outbreaks and learn from patterns too rare for any one company to detect alone. But AI depends on one thing: large, diverse datasets.
New research by Cornell doctoral candidate Linda Kalunga found that although food companies recognized clear benefits to pooling confidential food safety data, concerns about trust, competition and shared standards continue to stand in the way. The study, published in npj Science of Food, is based on interviews with 27 food industry executives, food safety directors and managers representing the dairy, meat, produce, food manufacturing and food safety laboratory sectors. Kalunga, a doctoral candidate in the lab of Renata Ivanek, professor in the College of Veterinary Medicine and co-director of Cornell Institute for Digital Agriculture (CIDA), led the study with collaborators from Cornell, the University of California, Davis, and the University of California, Berkeley.
Kalunga earned her Master of Public Health degree from Cornell in 2022 before beginning doctoral studies in Ivanek's lab. The study also brought together experts in food science, data science, economics and social science. "Before I began this research, I expected that companies would be hesitant to discuss sharing their food safety data, especially when it came to collaboration with competitors," Kalunga said.
"I was surprised by how openly participants shared their perspectives. Many were willing to discuss not only the broader industry-level hurdles, but also their own company's challenges, highlighting a shared recognition of the issue and a willingness to talk about potential solutions." Across interviews, participants consistently described the same pattern: They recognized the value of pooling data but struggled with technical barriers and a lack of trust that their data would be handled responsibly. "Sharing confidential food safety data with competitors to generate AI-driven insights from larger datasets is a tempting proposition," Ivanek said.
"It has enormous potential benefits, but also many ways to fail." Participants said larger shared datasets could help companies identify trends earlier, improve predictive models and better understand rare foodborne outbreaks. Several added that pooling data could also give smaller companies access to insights that would otherwise require costly investments in research and analytics. Together, those advantages could help companies detect food safety problems sooner and improve efforts to keep contaminated products out of the food supply.
However, companies cited inconsistent recordkeeping, incompatible data systems and uneven digital infrastructure as drawbacks. While larger firms often use sophisticated digital platforms, many smaller companies still rely on spreadsheets or paper records, making large-scale information sharing difficult. Kalunga said the findings pointed to trust, not technology, as the greatest obstacle to industry collaboration.
Extract — continue reading at the source.
