Societal Verification

With the NPT under increasing strain, the JCPOA defunct, and New START having expired in February 2026, formal verification mechanisms are steadily disappearing. For arms control verification—the process of collecting, analyzing, and checking data to help assess whether a state is fulfilling its obligations under an international agreement or treaty—knowledge production using national technical means (NTM) and contractually regulated inspection regimes continues to play a key role. However, analyses by independent non-state actors—including NGOs, think tanks, and independent analysts—using open-source intelligence (OSINT) have increasingly come to the fore, broadening the debate. This information has largely been disseminated through open-access reports and analysis, media statements, and other forms of public engagement concerning states such as North Korea, Iran, and China. While non-state actors do not officially verify arms control agreements or perform actual verification tasks, they do gather verification-relevant information and share it publicly. In democratic states, societal verification has enabled OSINT analysts outside the government–intelligence nexus to contribute to the dissemination of knowledge, using, for example, commercial satellite images, trade data, and shipping information. In doing so, they provide the public with information that complements government accounts, at times even revealing previously undisclosed developments, including Iranian reconstruction activities following U.S. and Israeli strikes and on-the-ground activities in China related to the construction of missile silos.

Broadening the availability of reliable information on state nuclear programs—particularly amid growing global power competition and in crisis situations—could raise public awareness, promote a more balanced narrative, and, at times, counter the tendency of citizens to assume worst-case scenarios when forming their opinions. It can also serve as a stepping stone toward mitigating nuclear risks and help pave the way for future formal arms control treaties. The monitoring of North Korean nuclear facilities and tracking of Iranian sites by non-state actors illustrate how satellite imagery analysis produced by NGOs feeds into public debate and media reporting, while also being used by government officials.¹ However, the availability of these OSINT analyses and the associated media coverage tells us little about their ultimate impact on policy debates or government decisions. This would be a worthwhile avenue for future research.

While nuclear verification serves as the empirical anchor in this contribution, the production of verification knowledge also extends to chemical weapons monitoring, environmental treaty compliance, and humanitarian crisis assessment—fields in which open-source analysis is playing an increasingly important role. Thus, the broader trends analyzed here are, to some extent at least, transferable to other verification domains.

Trajectory of a Concept

As formal treaty regimes erode, societal verification—an interdisciplinary approach that gives non-state actors, such as whistleblowers, scientists, and experts, a role in arms control verification by drawing on open-source intelligence²—has acquired increasing practical relevance. Remote open-source monitoring is no longer merely a supplement to official verification; it is increasingly providing valuable verification-relevant information. The literature on societal verification is conceptually rich³ but empirically thin. Existing studies establish the normative case for involving civil society in verification but fail to systematically map who is currently undertaking this work, what tools they are using, and under what conditions.

First articulated by Szilard⁴ and subsequently developed by Rotblat⁵ and others, societal verification posits that nongovernmental actors can and should contribute to arms control verification through OSINT. The objective of societal verification activities is to confirm a state’s compliance with its treaty obligations by relying on publicly accessible information. Early conceptual thoughts revolved around two models: “whistleblowing” and “civil society reporting”.⁶ Today, however, there is a general consensus that the focus has shifted from individual citizens to well-established expert communities performing societal verification tasks.

Opportunities

There is considerable potential for societal verification and the involvement of experts in detecting and demonstrating the absence of treaty violations in response to allegations against a state. Indeed, there is room for “(…) non-governmental organizations and individuals to engage in tasks associated with treaty verification, including monitoring compliance and supporting investigations with additional data”.⁷ Non-state OSINT analysts bring information into the public realm that, although not necessarily new to the intelligence community, can offer additional perspectives and assessments on conflicts and proliferation hot spots, complementing those released by the government. This helps to address crucial lacunae and raises citizens’ awareness. From the interviews conducted in March 2026, two points stand out that illuminate the impact of verification-relevant knowledge produced through OSINT-based assessments by non-state actors.

Informing the Public

While the intelligence community, by its very nature, does not publish its sensitive findings, open-source analysts do, thereby helping to keep the public informed. They also raise awareness on issues that might not otherwise be at the center of the broader public attention. Indeed, the importance of non-state OSINT-based analyses derives from them being publicly available and accessible to all. It is here that non-state actors bridge the gap by providing expert analysis, which, together with other sources of information—such as journalism, government press releases, and social media—helps to create a more complete picture. This is highly beneficial, as increasing the information available to the public enables them to make more informed decisions, gain additional perspectives and context on particular policies, and develop a broader understanding of a conflict.

Thus, think tanks and expert NGOs have an important role to play here in disseminating informed analyses and providing fact-based, balanced (counter)narratives to those advocating drastic action based on threat-escalation scenarios. Yet the reverse can also occur: analysts may assess conflicts and threat scenarios as more serious than governments do and consequently advocate tougher measures. Although intelligence communities may have reached the same conclusions, perhaps even earlier than those monitoring developments through available OSINT data, current and former government employees acknowledge the importance of the work of think tanks and NGOs and regularly access, read, and draw on their publications. They regularly engage with and request briefings from non-state analysts. Until a year ago, Washington had established communication channels that were routinely used to bring together government officials with non-state OSINT analysts from NGOs and think tanks. However, with the recent cuts to U.S. State Department and think tank funding, the prospects of continuing collaboration look rather dim. This has been a direct and widely felt consequence since the start of Trump’s second term, lamented by government and independent actors alike.

Yet, U.S. State Department officials and International Atomic Energy Agency (IAEA) analysts continue to consult and engage with the work of non-state actors. One reason is pragmatic and relates to the types of information government officials are permitted to cite publicly. Here, NGOs and think tanks that track global nuclear arsenals, such as the Federation of American Scientists (FAS) and the Stockholm International Peace Research Institute (SIPRI) provide useful publicly accessible estimates. The Status of World Nuclear Forces⁸ published by SIPRI, for example, provides detailed figures on developments in military stockpiles. That said, even within government, discussions are necessary when reports and analyses by non-state actors emerge, as not everyone has the same level of insight or security clearance. Public reporting, on the other hand, is something everyone working inside government can read.⁹

Stimulating Domestic Debate

Reports and analyses by U.S. non-state actors are opening up domestic debate. If these actors publish something—for example, on Iran, Chinese silo fields, or the return of U.S. missiles to Lakenheath in the UK—in media outlets such as the New York Times or the Washington Post, this can spark a discussion among a range of stakeholders. While the intelligence community and some government agencies were already aware of China’s silo fields, it was only after the Federation of American Scientists published its findings that the issue entered the broader domestic debate.¹⁰ Analyses by non-state actors can bring issues to the fore that might otherwise have gone undiscussed, thereby prompting a much more open debate. In this particular case, it also had repercussions in China with the Chinese Foreign Affairs Ministry releasing a statement in response to the FAS numbers.¹¹ A non-state analyst’s work might resonate with different segments of the broader population, while their analyses and statements might also elicit a response from the government. Whether or not this always makes a useful contribution to curbing nuclear-related threats is debatable, however, since back-channel diplomatic efforts may be underway in parallel without public knowledge. In such cases, the untimely release of OSINT analysis by non-state actors can undermine bilateral efforts by turning public opinion against concessions to a state perceived to be in breach of an agreement. Yet, societal verification activities illuminate important topics that would otherwise, in all likelihood, have remained exclusively within the domain of classified government discourse. Although, as noted above, this may not always be beneficial—particularly with regard to the timing of OSINT releases—in democracies, it undoubtedly helps citizens to hold their respective governments accountable. This also concerns domestic issues, such as plutonium pit production in the U.S. and NGOs’ monitoring of the associated environmental impacts.¹² It thereby provides a domestic forum for open debate accessible to all. This is a highly salient point, as non-state OSINT analyses extend the conversation, previously confined to closed government circles, to a broader public, including voters. This, in turn, may indirectly influence the direction of policies in the making.

Limitations

In a convincing account, Al-Sayed sets out the potential asymmetries inherent in publicly available information focusing on verification, which was often instrumentalized to advance “US and Western strategic interests”.¹³ At the same time, publicizing such information risks reproducing preconceived normative stereotypes about countries perceived to be Western adversaries, such as North Korea, Iran, and China. Moreover, it could create a “hype” around specific issues that would subsequently consume a disproportionate share of public attention. Kretzschmar et al.¹⁴ thus caution that societal verification is far from unbiased and initiatives employing the concept should be aware of unintended consequences as the concept comes “with its own set of uncertainties”.¹⁵

Besides questions of interpretive authority, there are also practical limitations. Several authors stated that societal verification efforts rarely produce “smoking gun” evidence. As held by Hinderstein and Hartigan,¹⁶ in most instances, “(…) these tools are likely to convey information that is only indicative of potential violations or misuse” and can thus serve as no more than supplemental information. The concept, therefore, is best understood as complementary to state-level and international agency mechanisms, broadening the toolbox of available approaches and enhancing their capacity to alleviate suspicions and confirm violations.¹⁷

Lastly, there is a structural obstacle concerning the integration of the data produced into existing and traditional information streams for arms control verification. Currently, practitioners of the concept face a gap: they possess the tools to detect treaty violations, but the data they gather does not, at least officially, inform official government analyses. Put differently, NGO experts lack official protocols to publicize their observations without taking the public by surprise, releasing reports that could impact both bilateral and multilateral relations, potentially distorting the debate.

Outlook

Analyses drawing on non-state OSINT produced by expert communities and NGOs fulfill an important role. By disseminating information about developments surrounding global nuclear stockpiles, illicit construction activities, and the acquisition of sanctioned military equipment, they provide “(…) much needed transparency and data verification functions”.¹⁸ Open-source analysis enriches the domestic debate by providing citizens with new information that complements government accounts.

However, access to the satellite imagery that non-state analysts rely on in their work has been noticeably restricted. In March 2026, for instance, U.S. satellite imagery providers Planet Labs and Vanator imposed an indefinite blackout on high-resolution images of the Middle East. Providers announced that this blackout followed consultations with the U.S. Government and was intended to prevent information about U.S. military personnel in the region from falling into the hands of its adversaries. It remains to be seen whether these blackouts set a precedent.

In democratic societies, states must accept that information is continuously released to the public by actors outside the government. This is simply a reflection of the technical possibilities available today. What is also needed going forward is improved dialogue between the government and non-state actors using OSINT to broaden the debate, thereby enabling policymakers to make better use of information generated outside the official government–intelligence nexus. These two streams of information should not be seen as competing with one another, but ideally as complementary, working together for the benefit of the public. Moreover, OSINT is already used as one of several sources supporting the IAEA’s safeguards efforts and is permitted under Comprehensive Nuclear-Test-Ban Treaty (CTBT) rules. In sum, drawing on OSINT-based analyses and information produced outside the government–intelligence nexus within treaty regimes could, arguably, help address the current impasse in non-proliferation and arms control by broadening the discussion to include civil society perspectives.