Sustainable and Healthy Food Management System using Generative AI

Prosjektet «Bærekraftig og sunn mat ved bruk av generativ kunstig intelligens» (AI-SHeFMan) har som mål å støtte et bærekraftig matforbruk i Norge. Prosjektet setter søkelyset på sjømat og bruker generativ kunstig intelligens til å skape et matforvaltningssystem som hjelper forbrukerne med å ta sunnere og mer miljøvennlige matvalg. Systemet tilbyr personlige forslag til matretter og gir tilbakemeldinger om hvordan mat påvirker miljøet. Dette gjør det enklere for forbrukerne å innta bærekraftige vaner. Ved å samarbeide med industripartnere slår prosjektet sammen data om sjømatens bærekraft, kvalitet og ernæring. Dette er i tråd med Norges klimamål. AI-SHeFMan-prosjektet gir også verdifull innsikt for beslutningstakere ved å bidra til informerte beslutninger for å redusere Norges karbonfotavtrykk.

Norway’s seafood industry prioritizes environmental responsibility, evident in initiatives like the sustainability portal, but challenges persist when it comes to consumers, as they face misinformation and conflicting messages regarding food sustainability and health. To tackle these challenges, AI-SHeFMan aims to raise awareness using easily accessible personalized information on sustainability, quality, and safety of seafood throughout the value chain. AI-SHeFMan is committed to ethical AI practices, emphasizing data privacy, security, and responsible use of AI. We aim to mitigate biases in our AI models by ensuring they reflect the broad spectrum of Norway's population, including various ethnicities, religions, and genders. This approach is critical in personalizing dietary recommendations that are culturally sensitive and inclusive of all dietary practices, including those based on religious beliefs or personal choices such as vegetarianism or veganism. WP1 maps key stakeholders, collect/screen data for seafood products relevant to the project; nutrition, quality, safety, CO2 footprint, waste and by-products. WP2 evaluates the holistic environmental performance of studied seafood products (from WP1) throughout their entire value chain, using Life Cycle Assessment (LCA). WP3 develops comprehensive Norwegian meal databases by leveraging open data resources, to construct a set of meal guidelines that are infused with WP1 and WP2 data; LCA data, carbon footprint calculations and nutrient profiling; in support of the AI-powered mobile app in WP4. WP4 is the technical implementation and testing of AI-SHeFMan mobile app. AI-SHeFMan brings a unique blend of interdisciplinary R&D competencies in seafood quality, safety and nutrition, environmental and climate sciences, and health and data sciences. Our industry partners provide relevant seafood data, and they potentially benefit from new consumer insights from our mobile app.