The IAEA is exploring the use of artificial intelligence to strengthen food authenticity testing and combat food fraud. The project combines nuclear and isotopic analytical techniques with machine learning approaches to process large analytical datasets and improve prediction models for food origin verification. Initial benchmarking studies demonstrated that Decision Tree Classifier models can significantly improve rice origin identification while providing a cost-effective and scalable solution. Ongoing work will assess additional AI tools and platforms to support the authentication of a wider range of food commodities, contributing to stronger food safety systems, enhanced consumer confidence, and improved regulatory capacity in Member States.
Activity Type AI Tools/SolutionsResearch/Reports/AssessmentsThis project aims to promote and enable the use of isotopic techniques with AI tools for better management of water and environmental resources. It serves as a platform for scientists to share experiences and find synergies between isotope techniques and AI to inform policies.
Activity Type Research/Reports/AssessmentsAwareness/AdvocacyNetworks/Mentorship/Exchange
This project encompasses IAEA activities related to assistance to Member States in the area of AI and machine learning for nuclear security, addressing both opportunities and associated risks. The work focuses on the secure and trustworthy deployment of AI-enabled systems in areas such as computer security, radiation detection, data analytics, and threat detection.
Activities include coordinated research projects on enhancing computer security of AI applications for nuclear technologies, computer security for small modular reactors, and capacity building activities on AI for nuclear security. The project also supports international conferences, technical meetings and expert exchanges to strengthen understanding of vulnerabilities such as data manipulation, code insertion, model integrity and algorithm transparency. Through research, guidance development and capacity building, the project enhances Member State capabilities to apply AI in nuclear security while maintaining robustness, resilience and alignment with international nuclear security guidance.
This project assesses how machine learning on FPGA devices can help discriminate neutrons from gammas in a detector. The IAEA contributed measurements for the study, which aims to improve quantitative results and enable real-time discrimination.
Activity Type Research/Reports/AssessmentsAI Tools/Solutions