Nuclear Regulators Want Machines to Monitor Cyberattacks on Power Plants

Federal nuclear regulators want external researchers to demonstrate how artificial intelligence and machine learning can pinpoint cyberattacks against the nation’s nuclear power plants. 

“To prepare to regulate nuclear applications of AI/ML, the [Nuclear Regulatory Commission] plans to conduct research activities to develop insights and fundamental knowledge about AI/ML and the AI/ML use case,” NRC officials wrote in a recent federal contracting notice.

This request comes almost a year since the commission asked for feedback regarding how AI and ML technologies are used in nuclear power operations—and will be used down the line. 

For this latest work, the NRC wants to collaborate with an entity that can provide existing personnel, equipment and facilities to implement a test case for full assessment. 

“The research conducted by the vendor is expected to produce data that evaluates the impacts of AI/ML concepts, technologies and applications on nuclear power cybersecurity outcomes and programs, especially those outcomes and programs that may be relevant to new and advanced reactor designs,” officials wrote. 

A technical report documenting outcomes will be produced from the effort. 

NRC noted that potential partners must have a range of capabilities to meet their criteria, including assets to simulate nuclear power plants and cyberattacks on them, resources to complete the work in 16 months and more.

“If the vendor cannot demonstrate the level of formal training or experience specified above, they may submit a plan for how they will acquire the required training and experience,” officials wrote.

Responses to this sources sought notice are due Feb. 18.