Artificial intelligence for food safety

Artificial intelligence for food safety

A literature synthesis, real-world applications and regulatory frameworks

Ficha técnica

Número de páginas:
Idioma:
ISBN:
Año de edición:
Editorial y ciudad:
88
Inglés
978-92-5-140196-5
2025
SPE3, Valencia
Accede a la publicación

Artificial Intelligence (AI) is increasingly applied in food safety management, offering new capabilities in data analysis, predictive modelling, and risk-based decision-making. A review of the literature identifies three primary areas of application: scientific advice, inspection and border control, and operational activities of food safety competent authorities. Five country examples with the real-world use cases illustrate diverse uses of AI tools, including pathogen detection, import sampling prioritization, and language models for regulatory data processing. Regulatory frameworks, as well as voluntary governance, addressing AI in the public sector are emerging worldwide. National and international initiatives often highlight the importance of data governance, transparency, ethical considerations, and human oversight. Challenges such as biased data, explainability, and data governance gaps appear across different contexts, along with potential risks from deploying AI systems prematurely. Access to high-quality, interoperable data and collaboration among stakeholders can support effective integration of AI technologies. AI readiness often depends on understanding specific problems to be addressed, current capacities, and the quality of available data. Human oversight and continuous evaluation contribute to maintaining trust in AI systems. Collaborative efforts involving academia, the private sector, and international organizations help build shared knowledge and resources for AI development in food safety. Overall, AI presents opportunities to enhance resilience, efficiency, and responsiveness in food safety systems. Careful consideration of governance, data management, and multi-stakeholder cooperation can shape AI’s contribution to achieving sustainable and equitable outcomes in agrifood systems.

Índice

1. Introduction
 1.1. Background
 1.2. Relevance to food safety in the agrifood systems
 1.3. Purpose of the document and target audience
 1.4. Methods

2. Literature synthesis results
 2.1. Overview
 2.2. Applications of artificial intelligence in food safety management
 2.3. Algorithms

3. Artificial intelligence case studies for food safety management
 3.1. Overview
 3.2. Use cases of traditional and generative artificial intelligence
 3.3. Using machine learning to predict pathogen adaptation to food sources
 3.4. Import sampling prioritization with machine learning
 3.5. Proof-of-concept experimentation using language models for food safety
 3.6. Building human-centric artificial intelligence systems for emerging food safety risk identification

4. A global regulatory snapshot of artificial intelligence frameworks
 4.1. Responsible use of artificial intelligence within the public sector
 4.2. Example of preliminary activities conducted by authorities
 4.3. Global efforts and good practices
 4.4. International and multisectoral collaboration and partnership

5. Considerations for the use of artificial intelligence in food safety management
 5.1. Identify the problem first
 5.2. Value of the artificial intelligence tools
 5.3. Value of the artificial intelligence outputs
 5.4. Explainable artificial intelligence
 5.5. Possible pitfalls, challenges and risk management
 5.6. Data governance and data gaps
 5.7. Public algorithm sharing mechanisms
 5.8. Artificial intelligence literacy and capacity development
 5.9. Support for data-driven decision-making

6. Tips for food safety competent authorities
 6.1. Consider some key activities to be completed first
 6.2. Assess the current capacity for artificial intelligence development
 6.3. Ensure the readiness of data
 6.4. Step back and take a strong agrifood systems approach
 6.5. If the data is not ready, consider generating quality data for the long run
 6.6. Actively collaborate with various stakeholders for artificial intelligence development

7. Conclusions and the way forward

whatsapp
Plan de Recuperación, Transformación y Resiliencia Financiado por la Unión Europea