Enhancing Computer Security of Artificial Intelligence Applications for Nuclear Technologies
Closed for proposals
Project Type
Project Code
J02024CRP
2474Approved Date
Status
Description
Artificial Intelligence (AI) and Machine Learning (ML) applications are being increasingly deployed across the nuclear industry, from enhancing operational efficiency to strengthening security measures. These technologies offer significant benefits, including enhanced threat detection, automated monitoring, and improved incident response capabilities. As nuclear facilities adopt AI-enabled technologies, particularly in Small Modular Reactors (SMRs) and other new nuclear applications not holding to traditional designs, new computer security challenges emerge that require innovative approaches to preserve and protect functions important to safety and security. The integration of AI systems introduces unique vulnerabilities, including the potential for data manipulation during development or deployment, challenges in algorithm transparency, and the need to protect training data integrity.
The research activities planned for this project will assist with the development, use, and sustainment of technological capabilities related to the secure implementation and operation of AI systems in nuclear facilities by applying a systematic approach to identify key vulnerabilities, protection mechanisms, and security controls. It also enhances scientific research in AI security, encourages innovation in secure AI deployment, and increases the number of research and development professionals with expertise in both nuclear security and AI technologies.
Additionally, this CRP will develop AI-enabled security assessment tools and training frameworks for human resources development purposes, strengthening the capability of Member States to implement and maintain secure AI systems in nuclear applications, which in turn helps to ensure the effectiveness and sustainability of peaceful nuclear applications and ensures that the nuclear workforce that may otherwise be affected by the widespread deployment of AI-enabled technologies maintains continual engagement and opportunity.
This CRP increases international cooperation on and access to AI security technology and innovation and enhances knowledge sharing on mutually agreed terms, including through improved coordination among CRP participating organizations, particularly at the IAEA level. The establishment of sustainable networks of AI security experts and shared evaluation frameworks ensures long-term benefits for Member States in maintaining secure AI implementations in nuclear facilities are achieved in an open and inclusive manner.
Objectives
The overall objective of this CRP is to develop methodologies and frameworks to support the enhancement computer security within AI-enabled technologies intended for deployment within nuclear security regimes and strengthen existing computer security strategies and programmes supporting the adoption of AI-enabled technologies while protecting safety and security, enabling facilities and countries to become informed consumers of AI-enabled technologies.