Special report: AI Institute for next generation food systems (AIFS)

Authors: TAGKOPOULOS, ILIAS - University Of California, Davis; BROWN, S - University Of California, Davis; LIU, XI - University Of California, Davis; ZHAO, QING - Cornell University; ZOHDI, T - University Of California Berkeley; EARLES, J - University Of California, Davis; NITIN, NITIN - University Of California, Davis; RUNCIE, DANIEL - University Of California, Davis; Lemay, Danielle; SMITH, AARON - University Of California, Davis; RONALD, PAMELA - University Of California, Davis; FENG, HAO - University Of Illinois; YOUTSEY, GABRIEL - University Of California, Davis

Submitted to: Computers and Electronics in Agriculture
Publication Type: Review Article
Publication Acceptance Date: 2/24/2022
Publication Date: 3/18/2022
Citation: Tagkopoulos, I., Brown, S.F., Liu, X., Zhao, Q., Zohdi, T., Earles, J.M., Nitin, N., Runcie, D.E., Lemay, D.G., Smith, A.D., Ronald, P.C., Feng, H., Youtsey, G.D. 2022. Special report: AI Institute for next generation food systems (AIFS). Computers and Electronics in Agriculture. 196. Article 106819. https://doi.org/10.1016/j.compag.2022.106819.
DOI: https://doi.org/10.1016/j.compag.2022.106819

Interpretive Summary: Artificial Intelligence (AI) has the potential to transform US food systems by targeting its biggest challenges: improving food yield, quality, and nutrition, decreasing resource consumption, increasing safety and traceability, and eliminating food waste. Despite big leaps in AI capacity, food systems present several challenges in the application and adoption of AI: (1) Food systems are highly diverse and biologically complex, (2) ground-truth data is sparse, costly, and privately held, and (3) human decisions and preferences are intricately linked to every stage of food system supply chains. To address these challenges and transform U.S. food systems, the AI Institute for Next Generation Food Systems (AIFS) aims to develop AI technologies and nurture the next generation of talent to produce and distribute more high-quality nutritious food with fewer resources. AIFS has six research clusters, including two Foundational Research Areas (Use-Inspired and Foundational AI, and Socioeconomics and Ethics) and four Application Research Areas spanning the entire food supply chain: Molecular Breeding, Agricultural Production, Food Processing and Distribution, and Nutrition. We are developing generalizable, data efficient, and trustworthy AI solutions based on a knowledge-driven and human-in-the-loop learning paradigm designed to handle food system diversity and biological complexity, efficiently capture and utilize food system data, and garner user trust via explainability, safety, privacy, and fairness. Innovations in research are complemented by transformative and inclusive education and outreach approaches to nurture the next generation talent in a diverse workforce, and comprehensive plans to broaden societal engagement including knowledge transfer and collaboration.

Technical Abstract: Artificial Intelligence (AI) has the potential to transform US food systems by targeting its biggest challenges: improving food yield, quality, and nutrition, decreasing resource consumption, increasing safety and traceability, and eliminating food waste. Despite big leaps in AI capacity, food systems present several challenges in the application and adoption of AI: (1) Food systems are highly diverse and biologically complex, (2) ground-truth data is sparse, costly, and privately held, and (3) human decisions and preferences are intricately linked to every stage of food system supply chains. To address these challenges and transform U.S. food systems, the AI Institute for Next Generation Food Systems (AIFS) aims to develop AI technologies and nurture the next generation of talent to produce and distribute more high-quality nutritious food with fewer resources. AIFS has six research clusters, including two Foundational Research Areas (Use-Inspired and Foundational AI, and Socioeconomics and Ethics) and four Application Research Areas spanning the entire food supply chain: Molecular Breeding, Agricultural Production, Food Processing and Distribution, and Nutrition. We are developing generalizable, data efficient, and trustworthy AI solutions based on a knowledge-driven and human-in-the-loop learning paradigm designed to handle food system diversity and biological complexity, efficiently capture and utilize food system data, and garner user trust via explainability, safety, privacy, and fairness. Innovations in research are complemented by transformative and inclusive education and outreach approaches to nurture the next generation talent in a diverse workforce, and comprehensive plans to broaden societal engagement including knowledge transfer and collaboration.