INTRODUCTION

It is all too easy to imagine innumerable possibilities for nuclear theft—a parade of horrors. It is extremely difficult, however, to determine where the line should be drawn between credible and incredible risks, between risks that should be safeguarded against and those that can be safely ignored. An assessment of the risks of nuclear theft is even more speculative than an analysis of the risks of major accidents in the operation of nuclear power reactors. With respect to reactor operation, risks to public safety arise primarily from the possibilities of malfunctioning machines. In regard to nuclear theft, however, the risks to national and individual security arise primarily from malfunctioning people.

Nevertheless, the safety risk analysis applicable to reactor accidents and the analysis of security risks applicable to nuclear theft have two difficulties in common. In the first place, both types of analysis deal with very low probability risks of very great damage. It is noteworthy, however, that the damage which might result from a nuclear theft is potentially much greater than the damage that could result from the maximum credible accident in the operation of a nuclear power reactor. Second, as to both areas of risk, there is, and hopefully will continue to be, a lack of actual experience involving substantial damage to the public on which to base predictions.

As fuel for power reactors, nuclear weapon material will range in commercial value from $3,000 to $15,000 per kilogram—roughly comparable to the value of black market heroin. The same material might be hundreds of times more valuable to some group wanting a powerful means of destruction.