Assessment of IAEA Safeguards Implementation at the State Level! Troubles Ahead?

Initially, NPT safeguards focused on the verification of declared nuclear material, improving the IAEA’s ability to address undeclared nuclear activities has been a continuing driver for change. First, the IAEA recognized the need to address the risk of undeclared production of plutonium or HEU at declared facilities. It also developed a framework for implementing safeguards based on clear assumptions about diversion, and agreement on goals; how to meet them; and how to assess and convey goal attainment on a facility-by-facility basis. Events in Iraq and the DPRK in 1991 awakened the IAEA to the fact that undeclared nuclear programs could be carried out at locations out of its reach under routine circumstances. The development and adoption of the Model Additional Protocol in 1997 strengthened its ability to address undeclared nuclear activities at undeclared locations. This led to the development of “integrated safeguards” under which the IAEA moved away from the facility-by-facility approach and applied safeguards at the level of a State. Ground rules for timeliness and detection probabilities were changed, and the IAEA took advantage of new information technologies, the wealth of new information provided by States, and new internal processes to evaluate states’ nuclear programs and their internal consistency. Nonetheless, safeguards implementation relied on facility-based, implementation procedures. Now the IAEA is turning towards a more information driven implementation approach. It is to be: holistic, customized by State, based on all available safeguards-relevant information, and driven by State-specific factors. Are there “troubles” ahead? Maybe. Member States expect non-discrimination. How will this be conveyed in a “customized” world? How will acquisitions paths be prioritized and resources allocated? What assumptions are to be made? How will safeguards be evaluated and results conveyed? What state-specific factors are to be used? And how? Could IAEA assessments be rendered incorrect by States’ collaboration in technology and concealment methods? How can the IAEA avoid imparting over-confidence when the task is difficult?