Why the U.S. Intelligence Community Needs an OSINT Agency

Calls for a dedicated open-source intelligence (OSINT) agency in the intelligence community (IC) are hardly new. Over 20 years ago, two commissions were created to investigate, respectively, the surprise 9/11 terror attacks and the IC’s erroneous assessments of Iraqi weapons of mass destruction (WMD). Both found shortcomings in the IC’s exploitation of publicly available information. The 9/11 Commission recommended the creation of an OSINT agency while the WMD Commission faulted the absence of “any broader program to gather and organize the wealth of global information generated each day and increasingly available, if only temporarily, over the Internet.” 

The Office of the Director of National Intelligence (ODNI) was established in 2005 to better integrate the IC’s 17 elements. One of the first acts of the new director of national intelligence (DNI), John Negroponte, was to create an ODNI Open Source Center (OSC). But the establishment of the ODNI did not end turf battles within the IC, including over OSINT. The ODNI OSC was housed within the CIA and incorporated the CIA’s Foreign Broadcast Information Service, which had been translating foreign media reports since World War II. It is difficult to delineate the respective roles of the ODNI and the CIA in the IC’s OSINT effort over the following decade. In 2015, the CIA reasserted its leadership of OSINT, and, as part of a major CIA restructure, today’s CIA Open Source Enterprise (OSE) was established.

Criticism of the IC’s failure to make adequate use of OSINT has, however, only grown as the digital information revolution has accelerated. The comparative success of nongovernment OSINT organizations, particularly Bellingcat, has added weight to the critique voiced by many former intelligence officers, though at least one current intelligence officer has added a voice to this chorus. For example, a 2022 RAND study concluded that “open-source is not working, it is not getting better, and the Open-Source Enterprise had ample opportunity to change.” The RAND study was based on interviews with, among others, the recently departed principal deputy DNI, deputy DNI, chair of the National Intelligence Council, and the Defense Intelligence Agency director of analysis. 

The IC’s new OSINT Strategy 2024-2026 has been praised for getting the basics right, and current OSE Director Randy Nixon has been commendably forthright in explaining how the CIA is now adopting cutting-edge technology to meet the OSINT challenge. The 2024 strategy is titled “The INT of First Resort.” That is a modest goal, which was first articulated by the ODNI in 2006. The fact that it has taken the IC almost two decades to start addressing fundamental issues—such as establishing integrated OSINT collection management and common training standards—suggests that the IC’s suboptimal exploitation of OSINT is not simply the result of turf battles, bureaucratic structures, or technical capabilities. The IC’s culture of secrecy also constrains its embrace of OSINT. More fundamental reform is necessary to adapt the IC to the digital information age.

Fighting the Last War

There is a fundamental mismatch between IC institutions devoted to secret intelligence and the emerging information environment.

The IC was established to fight the Cold War. According to Mark Lowenthal, author of the seminal text Intelligence: From Secrets to Policy, during that contest “80 percent of the information the US required was secret and 20 percent was open.” It’s safe to assume that much more of the information Washington now needs can be openly or commercially sourced. 

The digital information revolution is generating a dramatically new operating environment for intelligence organizations. It’s not just the volume of data but also the diversity and speed. Everything from simple Google searches to the complex extraction of data from an increasingly fragmented cyberspace can provide valuable information. Private-sector intelligence capabilities increasingly match or exceed those of nation-states while commercial data brokers gather digital dust, which “includes information on nearly everyone that is of a type and level of sensitivity that historically could have been obtained, if at all, only through targeted … collection.”

The digital revolution is transforming not only the operating environment for the IC but also its purpose. Information is not only a means for understanding geopolitical competition but a central domain of that contest. While states have always competed to obtain and protect data, there is a growing need to be the fastest to gain actionable insights and act on them. Successful statecraft increasingly demands winning the information competition and shaping public narratives. The IC must further adapt to support this statecraft. 

Strategic Declassification—Innovation or Catch Up?

The IC has responded to this challenge by accelerating “strategic declassification,” defined by CIA Director Bill Burns as “the intentional public disclosure of certain secrets to undercut rivals and rally allies.” In the lead-up to Russia’s 2022 invasion of Ukraine, the U.S. and U.K. wrong-footed Russian President Vladimir Putin, pre-bunked Moscow’s disinformation, and accelerated the formation of a countervailing coalition by downgrading and declassifying secret intelligence. 

The process for strategic declassification has been refined since then. Whereas the ODNI once received one or two requests per month to lower the classification on reports, it now sometimes receives many more than that in a day. To obscure sources and methods, downgraded intelligence is disseminated in packages that combine secret intelligence and OSINT. 

Strategic declassification has been rightly hailed as a success. It has enabled the U.S. to better inform the public, counter disinformation, and shape narratives. Yet the underlying idea—that intelligence should be produced in forms that are optimally useful for policymakers—should not be novel. The piecemeal approach to declassification won’t keep pace with the accelerating demands of information competition. It’s also reasonable to ask how often the vaunted downgrade process is, in practice, simply applying the correct classification. Chris Rasmussen rightly asks why the military staging maps of Russian equipment shared with the Washington Post prior to the invasion of Ukraine needed to be classified in the first place. 

The tendencies of intelligence organizations to overclassify and undershare information are manifestations of a wider culture of secrecy. IC culture is the most frequently invoked (but rarely defined) obstacle to the IC’s embrace of OSINT.

It’s not surprising that the IC has fostered a culture of secrecy—it supports the mission of obtaining, protecting, analyzing, and disseminating secrets. But it has also inadvertently encouraged the view that information is valuable because it is secret rather than vice versa. Psychological factors such as the “endowment effect” and the “sunk cost fallacy” have probably reinforced that tendency.

Perennial problems like overclassification won’t be fixed through either ad hoc declassification or exhortations from CIA Director Burns to judiciously resist the “reflexive urge to keep everything classified.” Rather, new structural incentives are needed for the IC to more proactively produce more usable intelligence products.

An OSINT Agency for the Information Age

The establishment of a dedicated OSINT agency would be a step toward reconfiguring the IC for the challenges of the information age. Properly implemented, this reform should deliver improved OSINT and secret intelligence while optimizing the relationship between the two disciplines.

The most common argument for establishing an OSINT agency is that this is the only way to ensure that the discipline of OSINT obtains sufficient resources and heft within the hypercompetitive IC. For example, Amy Zegart argues that that “as long as open-source intelligence remains embedded in secret agencies that value clandestine information above all, it will languish” because those agencies won’t prioritize or champion OSINT. But why shouldn’t the IC just depend on the private sector to meet its OSINT needs? A growing array of commercial organizations now produce OSINT to supplement the work of traditional media and not-for-profit organizations like Bellingcat. But even the most reputable OSINT organizations are ultimately guided by their own interests, commercial or otherwise. An OSINT agency would have distinct authorities and be unambiguously guided by IC priorities. It would work closely with nongovernment OSINT providers and, unencumbered by the bureaucracy of secrecy, would be better able to do so. The OSINT agency’s seal of approval might be the only value it adds to a commercial product, but in an increasingly uncertain information environment, that reassurance will become ever more valuable.

A dedicated OSINT agency would also improve the production of secret intelligence. The accuracy of U.S. and U.K. insights into Putin’s intentions in Ukraine demonstrated the perennial need for and value of secret intelligence. But prioritizing scarce collection resources is a constant challenge. Isolating the most valuable secrets has become harder in the information age. The improved production of OSINT would avoid duplication and allow classified resources to be more sharply focused on hard targets. As the RAND report points out, “[U]ntil the IC truly invests in OSINT across the range of intelligence topic areas, the community will not know for sure which topics can be covered adequately by OSINT.”

The strongest argument against establishing a separate OSINT organization is that it runs directly counter to the goal of intelligence integration. Integrating OSINT and secret intelligence is an important objective, but the downsides of attempting to do so within a secret umbrella have now been amply demonstrated. OSINT differs fundamentally from secret intelligence precisely because it is derived from open sources. Indeed, OSINT differs more from secret intelligence than human intelligence, signals intelligence, and geospatial intelligence—each of which has a dedicated agency—differ from one another. IC leaders have long urged their officers to treat OSINT as the intelligence of first resort. A dedicated OSINT agency would remove the choice.

The challenge is not simply to integrate OSINT and secret intelligence but to optimize the relationship between the two disciplines. Productive competition between OSINT and secret intelligence should be enabled and encouraged. OSINT is faster, cheaper, and more easily shared than secret intelligence. Competitive pressure from improved OSINT should be used both to encourage secret agencies to target valuable secrets and to discourage them from overclassifying information. The output from a dedicated OSINT agency would create stronger incentives for the IC to proactively produce intelligence that is optimally useful for policymakers, rather than waiting for requests to downgrade and declassify.

The culture of secrecy has long constrained the IC’s embrace of OSINT. The power of OSINT should now be harnessed to disrupt the culture of secrecy. It’s distinctly possible that, over time, secret intelligence will become a niche specialization within a larger OSINT enterprise. Establishing a dedicated OSINT agency would not bring about such a radical change, but it would make it possible and leave the IC better postured for the information age.