CIA's Burns Spurs Spy Agencies to Embrace Book Learning, Data Extraction
Open Source Intelligence, long second fiddle in an espionage-first culture, seems finally to be getting its day, with CIA chief Burns leading the way
Although largely overlooked in spy thrillers, Open Source Intelligence, or OSINT—vacuuming up publicly available reports, books, and data—remains an enduring and high value component of any spy organization. Perhaps we can blame James Bond, who only read The Times of London, for OSINT’s perceived second class standing among the intelligence disciplines. Though to be fair, the peripatetic 1970s-era conspiracy spectacle Three Days of the Condor, despite starring a young and bookish Robert Redford, clearly did little to burnish the reputation of open source.
“I am not a spy. I just read books,” Redford’s CIA character, an open source practitioner, bemoans, wondering why on Earth somebody wants to kill him. And to be blunt, he does seem rather bland alongside Max von Sydow’s cold and dapper assassin.
Although there is no denying espionage and covert action possess a certain cloak-and-dagger charm in their mystery and clandestine derring-do, open source has a long history of intelligence successes. And as the recently released report, Open Source Strategy for 2024 -2026, promises, the role of OSINT has a bright future in U.S. intelligence. Prepared by the Office of the Director of National Intelligence and Central Intelligence Agency, the report outlines a strategy that includes what could be described as an aggressive expansion of effort and capability with CIA Director William J. Burns serving as “Functional Manager” of the campaign.
The way forward, as Burns explained it, is a distributed and cooperative effort spanning the entirety of the IC to include all 18 intelligence organizations. “As the IC’s Functional Manager for OSINT, I know the critical role that OSINT plays in defending our country and values,” he said in a prepared, albeit boilerplate statement. “In this pivotal moment, when OSINT is increasingly important and growing in demand, an IC-wide OSINT strategy is key to helping the IC move forward in a coordinated and determined way."
We’ve heard this for some time now. Fact is, the spy agencies have always valued “intelligence gleaned from highly classified sources, rather than information that’s openly available on the internet or elsewhere,” as the Federal News Network put it last August. “Right now, our customer really believes that the only valuable content that the IC produces comes from super classified sources,” Ellen McCarthy, a former chief of the State Department’s Bureau of Intelligence and Research, was quoted as saying, “and that’s just frankly not the case.” Gathering and making sense of, say, rumors of violent protests by rural Chinese factory workers, gleaned from provincial news reports by a skillful Mandarin-reading CIA analyst, can be just as valuable.
U.S. intelligence agencies seem to be finally coming around to that view. The strategic focus described in the report includes four areas for action:
*Coordinate open source data acquisition and expand sharing;
*Establish integrated open source collection management;
*Drive OSINT innovation to deliver new capabilities;
*Develop the next-generation OSINT workforce and tradecraft.
That is to say, the IC will develop new procedures and expertise to harvest the oceans of available data with tools like Artificial Intelligence or AI while keeping pace with advances in the field. Not addressed in the report, however, is speculation on how the new technology will change its consumers’ expectations. It is a well-worn though true trope that technology changes expectations of those who use it. Those of a certain age will no doubt recall the changes wrought by the proliferation of fax machines or email, not to mention smartphones.
Brave News World
Back in the day, before the advent of internet data, OSINT meant tracking foreign newspapers, radio reports, press releases, broadcasts, professional journals, and the like. Most often this effort yielded broad strokes, tidbits, and trends, but occasionally the golden nugget popped up.
In October of 1939, for example, an attentive German spy in New York City, Simon Koedel, clipped an Associated Press story that had been quietly buried on page 47 of The New York Times, headlined, “Roosevelt Protected In Talks To Envoys by Radio ‘Scrambling’ to Foil Spies Abroad.” Although the story, quoting some incautious engineers, gave scant technical details, Koedel spirited it to Bremen, and it wasn’t long before Reich intelligence was able to eavesdrop on the scrambled, frequency-hopping calls from a listening post in occupied Holland.
Today, sources of information have expanded from newspapers, magazines, and radio broadcasts to include an ever-growing number of internet sources, commercially available databases, and other large data pools. Given this wide ranging and evolving OSINT environment, the IC is now working to accelerate its approach when it comes to collecting and delivering high-value intelligence to policymakers and warfighters.
One significant element of this strategy will be the use of advanced and rapidly evolving technology, including tools like Generative Artificial Intelligence, which creates text, images, and even video from data. “The growth of generative artificial intelligence presents both opportunities and risks for OSINT tradecraft,” the report explained. “GAI can be a powerful tool to enable timely and insightful OSINT production, including by aiding the identification of common themes or patterns in underlying data and quickly summarizing large amounts of text.”
However, the use of new technology arrives with significant caveats. The report cautions against an over-dependence of such sophisticated tools and need for specialized training of personnel to mitigate risks, which include inaccuracies and “hallucinations,” in AI generated products. That is to say, AI tradecraft will include, if not heavily rely on, human-machine teaming. As some AI experts have noted, this arrangement is closer to a true collaborative relationship than simple data entry. Indeed, a Harvard Business Review article went so far as to label this process “collaborative intelligence.” As the article noted, “Through such collaborative intelligence, humans and AI actively enhance each other’s complementary strengths…” Translation: The ability to sift through and analyze terabytes of data is all fine and good but still requires human judgment and guidance.
Additionally, the IC’s new OSINT strategy envisions an increased emphasis on engagement with academia, the private sector, and foreign counterparts. Although such engagement is not new to the intelligence field—spies have traditionally been early adopters and adapters of new technology—it would seem particularly important in the case of AI. Not only is the technology evolving very quickly, but those innovations are occurring in fields as diverse as advanced scientific research and streamlining basic business practices.
What’s Old is New Again.
Neither the value nor technological challenges of OSINT are new to the current state of American intelligence. It is, arguably, the oldest component of the CIA. Predating the agency itself, the wartime Foreign Broadcast Monitoring Service was established in 1941 under the Federal Communications Commission before changing its name months after the Pearl Harbor attack to the more accurate Foreign Broadcast Intelligence Service, or FBIS.
The organization was originally created to identify, monitor, and analyze overseas broadcasts, including news reports and all shades of Axis propaganda transmitted on the relatively new and far-reaching technology of shortwave radio. Within just a few months of starting operations, the organization established a proprietary teletype line to the Coordinator of Information (predecessor of the Office of Strategic Services, or OSS) and shortly thereafter added another line into the Department of State.
By the war’s end, the FBIS had monitoring stations across the U.S. while its “wire service” had expanded in both scope and output, transmitting some 25,000 words a day to 18 defense offices, including the Office of War Information War Department and armed forces. Significantly, it had also forged ties with its allies’ counterparts, such as Britain’s BBC, in an early fruitful instance of the “special relationship.”
One example of a monitored wartime broadcast was Nazi Germany’s Radio DEBUNK (sometimes rendered D.E.B.U.N.K.) that claimed to be “The Voice of All Free America” broadcasting from America’s Midwest heartland while actually transmitting from Breman. Among DEBUNK’s more colorful tales were reports of rampant prostitution in England with hordes of fallen women fueling an epidemic of STDs among American servicemen, while President Roosevelt was portrayed as traitorously colluding with “Jewish communists” and “profiteers.” Then labeled “black propaganda,” today DEBUNK’s broadcasts would be called “disinformation.”
Another notable Nazi propagandist monitored by the FBIS was American-born Mildred Gillars, popularly known as Axis Sally. A failed model and actress, she was among that small number of Americans eager to sell their American accents and homespun colloquialisms to the Reich. In broadcasts from Berlin, she frequently addressed American women with virulently anti-Semitic, anti-British, and anti-Roosevelt diatribes.
“As one American to another—do you love the British? Well of course the answer is no,” Gillars asked during one broadcast. “Do the British love us? Of course—I should say not. But we are fighting for them. We are shedding our good, young blood for this kike war, for this British war—oh girls, why don’t you wake up...I love America—but I do not love Roosevelt and all of his kike boyfriends who have thrown us up into this awful turmoil.”
No doubt, this was ugly stuff. It played on newfound fears and long-held prejudices— cornerstones of the Nazi regime itself—though other broadcasts monitored by the FBIS would prove to have strategic value. In one instance, Berlin radio announced the bombing of a fortified site on the outskirts of Leningrad, which it had previously heralded captured. Washington was surprised: Analysis indicated Russian troops had recaptured the location without informing their allies. In another instance, monitoring of foreign broadcasts disclosed a shortage of Japanese transportation.
The FBIS staff of linguists and analysts, many with backgrounds far outside the norm of U.S. intelligence or military personnel, comprised what a reporter from Collier’s magazine described as “the greatest collection of individualists, international rolling stones and slightly batty geniuses ever gathered together in one organization.” One Serbian translator wore a beret (a fact that seemed exotic to the reporter) and was said to speak more than 20 languages. Another staffer, Rueben Fine, was a frequently married Bronx-born chess master. They affectionately dubbed themselves the “Screwball Division.” However, the self-deprecating humor was belied by the sophistication of then state-of-the-art technology and innovative techniques of “radio frequencies that could “cruise” for enemy broadcasts for changes in broadcast schedules along with any new transmissions that may pop up. And, too, much of the radio expertise was drawn from equipment manufacturers and commercial broadcasters, as well as the passionate community of ham radio enthusiasts.
The Office of War Information, headed by Pulitzer Prize winning playwright Robert E. Sherwood, was a particularly enthusiastic consumer of the FBIS product. Having the latest misinformation and disinformation talking points allowed him and the organization to counter them in a timely manner, both at home and overseas. At a 1944 congressional hearing, Sherwood said that following the Allied victory in Sicily, more than 80 percent of the German and Italian prisoners captured said that propaganda leaflets and broadcasts influenced their surrender.
After the Fall
Following the war, the service became a “founding component” of the CIA. On December 12, 1947, just a few months after President Truman signed the National Security Act of 1947 into law, the National Security Council issued Directive No. 6, aka the “FBIS Charter,” which authorized the Director of Central Intelligence to “…conduct all Federal monitoring of foreign propaganda and press broadcasts required for the collection of intelligence information to meet the needs of all Departments and agencies in connection with National Security.”
In some cases, this was easier said than done. The Soviets had a morbid fear of foreign intelligence-gathering bordering on the comical and frequently crossing the line into the absurd. For instance, accurate maps were considered state secrets from Josef Stalin’s ascendence onward, pre-dating the Cold War by decades. With cartographers under tight control of the security services, Soviet distortions included inaccuracies of scale, misplacement of roads and rivers, and exclusion of entire cities as well as passenger rail lines from public maps. The most accurate map of Moscow was said to be the one produced by the CIA. Remarkably, this policy endured long after the advent of satellite surveillance, which rendered the deceptions irrelevant against perceived adversaries, and wasn’t relaxed until the Gorbachev era.
Going Deep
Explicitly stated in the new report is that the capabilities of advanced tools such as AI, to wrangle immense amounts of open source data into viable intelligence products holds the potential for increased value across the entire IC.
“As the open source environment continues to expand and evolve at breakneck speed,” the report reads, “the ability to extract actionable insights from vast amounts of open source data will only increase in importance.”
Given the unique speed of change in the OSINT environment, the plan calls for annual strategy reviews with an eye to guiding implementation efforts. As the report noted, the speed of innovation is a critical measure of success, while the ability to test new capabilities on unclassified data systems presents fewer risks and barriers than classified networks.
“The IC OSINT Strategy,” Director of National Intelligence Avril Haines said in a prepared statement, “represents the beginning of a long-term process that will professionalize the OSINT discipline, transform intelligence analysis and production, and create new avenues for partnering with brilliant American innovators and like-minded foreign partners.”
It’s a worthy, even imperative, goal. Breaking down old habits is the real challenge.
On behalf of the Attorney General I was assigned to review all classified all source documents concerning Nazi war crimes committed in Belarus. My impression is that 80 percent of classified information can be obtained from open sources but one must have professional advice on where to look. Later I was required to submit my book manuscripts to CIA pre-publication review. I frequently needed open sources to justify writing about still classified topics. I had the great good fortune to have several mentors in the intelligence community who pointed me towards obscure books and archives. Lexus/Nexus digital archives of newspaper and magazines are almost as good as FBIS but quite expensive. Law students, however, have free access to this search service.
Atty. John J. Loftus (retired)
Where does the illustration come from? The types of information collection in it are textual, as you describe in the article, FBIS, news reports, etc. . But that's not the current "OSINT" journalists get sent to training for these days --- for identifying locations from IPhone photos and identification of weapons from serial numbers -- "Visual Information."
There seem to be more and more definitions of OSINT these days. I assume the intelligence community was the originator. It seems to have gotten away from them.
.