Last fall, I visited the Nevada National Security Site (NNSS—formerly known as the Nevada Test Site), a sprawling 1,300-acre complex in the picturesque desert and foothills outside Las Vegas. The NNSS once hosted facilities supporting Cold War nuclear testing and development programs. Veterans of Los Alamos and Lawrence Livermore National Labs’ testing programs, some of whom began working for the testing program in the 1950s, accompanied us on tours of modern and historic facilities. As I listened to these scientists discuss the nuclear physics principles they discovered, several questions important to U.S. nuclear policy came to mind: Does the United States still retain the capability to test a nuclear weapon 24 years since we self-restricted ourselves from doing so? Does the expertise to do so still exist in the nuclear security enterprise? What type of international incident or technical surprise would require something as drastic as a below- (or above-) ground test of the U.S. nuclear stockpile? Today, the NNSS is home to experimental work that, in some cases, cannot be done anywhere else in the world. The work aids some of the nation’s most important national security missions, including stockpile stewardship, nonproliferation, and counterterrorism. In many ways, U.S. nuclear security enterprise advances in supercomputing, analytical simulations, and other science-based stockpile stewardship programs have helped national lab scientists understand the science behind nuclear weapons more than the atmospheric, underground, and exo-atmospheric testing conducted during the Cold War. On the other hand, what if the United States needed to restart its testing program to evaluate its stockpile and maintain its strategic deterrent, especially considering much of the data we use to validate stockpile stewardship models comes from the Cold War-era testing program? I am not overly concerned about the difficulty of conducting an explosive test of one of our stockpile systems, either to send a signal to an adversary or to demonstrate resolve—nor am I advocating for a test in today’s security environment. A nuclear explosive test would impose a tremendous cost on the nonproliferation regime the United States and its partners have worked to establish over decades, but it would simply require time and resources to ensure that the test was conducted safely and created minimal radiological contamination. The time and resources required for a demonstration detonation pale in comparison to the time, resources, and expertisethat a test with modern diagnostics tools that yields useful data would require. In short, it is easy to verify that a weapon works. Because of the complex physics involved, however, understanding why a weapon works requires a much better understanding of what happens inside the weapon. To obtain the relevant data, a tremendous number of sensors, wires, and support equipment are required that can measure very small particles with high fidelity over an incredibly short period of time; something that is not achieved easily or cheaply. That the national labs were able to design test diagnostic tools in an era when computers had less processing power than an iPhone remains one of this country’s most impressive scientific achievements. Those developments, made by dedicated scientists in a dusty desert, have underpinned our nuclear stockpile and nonproliferation strategies for decades. The nuclear testing program virtually created its own industry of boutique sensors and equipment that could produce useful data, despite austere environmental conditions (those inherent to the Nevada desert and those created by a nuclear detonation). In a fitting metaphor for the overall state of our nation’s nuclear infrastructure, drill bits, pipes, and heavy-duty specially-made wires once used for the underground test program now lie rusting in the Nevada desert. On some of the antiquated equipment, much of which hasn’t moved since the United States’ self-imposed test moratorium in 1992, sits an optimistic sign that reads: “Test Readiness Program—We’re ready when you are!” As the equipment the national labs used ages in the Nevada desert, the remaining experts who knew what to do with that equipment have retired or moved on to work on other scientific issues. Very few are left in our national lab community who participated in the testing program, and even fewer are left who are not partially or fully retired. Every year, the directors of Lawrence Livermore, Sandia, and Los Alamos National Laboratories write a letter to the president certifying the nuclear stockpile and discussing issues such as test readiness and workforce demographics. In a testament to the ingenuity of the national labs and the success of the Stockpile Stewardship Program, none of these evaluations has expressed a need to return to testing or revealed a major issue in a stockpile system. But as the number of weapons in our stockpile decreases due to policy decisions and treaties negotiated at the highest levels, the need has never been greater to ensure that our stockpile systems always work in the unfortunate event they are needed, and to ensure that they never result in an unintended accident. Several factors will inevitably require the United States to revisit policies that have guided its nuclear activities since the end of the Cold War, including today’s international security environment, other countries’ nuclear modernization (and potential stockpile expansion), and North Korea’s impending missile threat to the contiguous United States. I am confident the current U.S. stockpile serves as an effective deterrent to potential adversaries against a nuclear attack on the U.S. homeland. I am less confident, however, that the stockpile will be sufficient to ensure deterrence in the mid- to late-future or to effectively provide extended deterrence to U.S. allies, especially in the case of technical surprise or if a new nuclear capability is needed. After the Cold War, the United States had unmatched conventional superiority, and the Russians were willing to negotiate treaties that reduced warhead numbers and restricted certain types of delivery systems. Unfortunately, today neither of these lynchpins of international security is likely to hold true, at least in the near term. In an increasingly complex and ambiguous international security environment, it has never been more important that we ensure that the nuclear security infrastructure—both the physical infrastructure and, perhaps more importantly, the professionals within the nuclear security enterprise—is up for the challenges the future may hold.