Food Assistant for Consumer Behaviour Change through Citizen Science and AI

The contemporary food system within the European Union is characterised by significant sustainability and health challenges. This paper presents an innovative approach to address these issues by leveraging Citizen Science (CS) and Artificial Intelligence (AI). The study involved thirteen participants from Spain, France, and the USA who shared pictures of their food receipts over a month through WhatsApp. The receipts were analysed using optical character recognition (OCR) technology, and the data obtained was used to generate tailored nutritional and sustainable living recommendations through a Large Language Model (ChatGPT v4.0). The key objectives were to foster healthier and more environmentally friendly shopping habits and to engage consumers in a scientific process by making them active participants in data collection. The findings indicate that the integration of AI-generated advice can positively influence consumer behaviour towards more sustainable and healthy food choices. The study also assessed participant satisfaction with the AI recommendations and their self-perception in assigned consumer clusters, such as Health-Focused and Sustainability-Conscious. The majority of participants reported a high level of satisfaction with the intervention and agreed with their cluster assignment, indicating the potential of AI tools in consumer health and environmental education. This paper provides open access to the collected data and the methodologies used, to benefit the public and facilitate further research.