This is a CBSI and the world. I'm John Bachelor. Gordon Chang at Gordon G. Chang, my colleague and co-host, nuclear weapons in the 21st century, an article at the Foreign Affairs magazine,

prestigious publication for the men and women who practiced foreign affairs in Washington in and out of administrations, teachers,

professors, think tankers, and secretaries of defense, for example. That Foreign Affairs magazine publishes an article by

Robert O'Brien, the National Security Advisor for President Trump at the end of

his term. Mr. O'Brien is often mentioned as a cabinet member and a potential second

Trump term. In it he says many things about what needs to be done for the defense industry and

on nuclear weapons he says

not only does the united states need new nuclear weapons new designs

but also those two designs must be tested as well as designed in computer models. We welcome

Peter Hughesi, president of geostrategic analysis, and also a fellow at the National Institute for Deterrence Studies.

Peter, I'm unfamiliar with the language.

Computer modeling is one understanding of how you build nuclear weapons.

Testing is another understanding.

Do we need both?

And what are our adversaries doing right now?

Good evening to you.

Good evening, John.

Our adversaries in 1999, Russia, decided to build small nuclear

weapons and did a lot of very small underground nuclear testing and the Elson's decree was build me small battlefield nuclear weapons that I can use to fight.

And the Chinese followed suit around 2009 and 8 and 9 and 10 during the great financial problem and then said

we're also going to build a huge nuclear capability. There's a debate in the community,

computers work, but they're not adequate or they are or what you get from

...