DoE on Verifiable Dismantlement of Nuclear Warheads
In anticipation of future nuclear arms control agreements that would require the dismantlement of nuclear warheads, the Department of Energy undertook a technical study during the Clinton Administration to determine how such dismantlement could be verifiably accomplished. The resulting report, experts say, is still the best available treatment of the subject.
A copy of the unclassified report, marked “official use only,” was obtained by Secrecy News and posted online today.
The DOE authors identified ten types of activities that could be used in a warhead dismantlement regime, involving various forms of monitoring at successive stages of the process. One or more of the ten could be employed, depending on the degree of confidence desired.
In principle, it should be fairly straightforward to dismantle a given nuclear warhead with confidence. However, “determining that an item to be dismantled is actually a nuclear warhead is very difficult” without compromising classified information, the report states. The use of x-rays or radiographs to confirm that an object is in fact a warhead “would be highly intrusive and would reveal highly classified nuclear warhead design information” to foreign inspectors, potentially exposing design vulnerabilities and other sensitive information. Such concerns might be addressed by other forms of monitoring, the report says.
The study concluded that “transparency measures for monitoring warhead dismantlement can be applied… with up to a moderate level of confidence that dismantlement has taken place if implemented at the Unclassified to [Confidential] level.” Verification that an actual weapon has been dismantled — which is a more demanding standard than mere “transparency” — can be achieved with an appropriate exchange of classified nuclear weapons design information.
The report provides a detailed description of the dismantlement process, a summary of previous dismantlement studies (including one by the Federation of American Scientists and another by the JASONs, but not the 1960s-era Project Cloud Gap study), and other valuable information that could serve to inform and accelerate current analyses of nuclear warhead dismantlement.
See “Transparency and Verification Options: An Initial Analysis of Approaches for Monitoring Warhead Dismantlement,” prepared by the Department of Energy Office of Arms Control and Nonproliferation, May 19, 1997.
Americans are paying too much for almost everything, because the United States has long treated its trucking industry as an artifact to be preserved rather than as an opportunity for innovation.
These ideas aim to advance the detailed policy solutions needed to foster public trust and implement fairness in the adoption of AI across diverse domains, from healthcare and government benefits to rural access, education, and worker protections.
The evidence is clear: algorithmic pay-setting is established in app-based work, and payroll/timekeeping failures show how software can produce systemic wage harm at scale
While a few states have taken steps to implement decision-making mechanisms for certain AI systems, too many leaders are simply accepting narratives about AI’s purported public benefit at face value – jumping to the “how” of AI implementation before thoroughly vetting potential systems and deciding whether they are appropriate to use at all.