Process examines illicit nuclear trade exposure

A Center for Advanced Defense Studies (C4ADS) and Nuclear Threat Initiative (NTI) report has determined a blueprint for identifying high-risk or illicit nuclear trade activities at scale.

Signals in the Noise: Preventing Nuclear Proliferation with Machine Learning & Publicly Available Information outlines how utilizing machine learning techniques analyzing commercially available trade data enable analysts to uncover previously unknown companies constituting a nuclear proliferation risk from millions of trade transactions.

“Illicit trafficking of nuclear materials and technologies around the world—whether by terrorist organizations, rogue states, criminal enterprises, or even unwitting mules—poses a serious threat to global security,” NTI Co-Chair and CEO Ernest J. Moniz wrote in the report’s foreword. “Those who engage in such criminal acts evade detection by operating in the shadows and in plain sight. New tools can help prevent nuclear proliferation by revealing footprints left by bad actors and to monitor and verify future arms control and export control agreements.”

The work showed automated data preparation and analysis could save hundreds of analyst hours and identify twice as many potential high-risk entities as previous efforts had through more manual approaches.

Recommendations to global leaders of nuclear non-proliferation efforts include integrating publicly available information more deeply into existing monitoring and verification regimes; using modern analytical approaches, including machine learning, to enable using big data at scale; building partnerships to allow analysts to access shared data; and embracing the use of publicly available information and modern analytical tools for future international non-proliferation initiatives.