Machine learning can help expose illicit nuclear trade, says new report

The main takeaway: According to the report, NTI and C4ADS worked together for two years on a pilot project to demonstrate the viability of this new approach to nuclear nonproliferation efforts. “The project succeeded,” the authors write in the report’s executive summary. “Trade network analysis—and the machine learning processes that supported it—uncovered previously unknown entities of elevated risk within millions of transactions. The work showed that automated data preparation could save hundreds of analyst hours and help identify twice as many potentially high-risk entities as previous manual efforts.”

Recommendations: The report calls on the leaders of nonproliferation efforts around the world to do the following:

■ Integrate publicly available information more deeply into existing monitoring and verification regimes.

■ Use modern analytical approaches, including machine learning, to enable using big data at scale.

■ Build partnerships to allow analysts to access shared data.

■ Embrace the use of publicly available information and modern analytical tools for future international nonproliferation initiatives.