
In the early 1980s, the Air Force Office of Special Investigations ran a deliberate, documented disinformation campaign against a private American citizen who had picked up something real near Kirtland Air Force Base. He was not an enemy agent. He was a civilian electronics engineer. They destroyed him anyway, and the playbook they used is the clearest documented example of what the apparatus does when it decides information must be controlled.
Albuquerque, New Mexico. Late 1970s. A physicist picks up something he cannot explain.
Paul Bennewitz was not a crank. He held a doctorate in physics, ran a successful electronics company called Thunder Scientific Corporation, and lived in a house adjacent to Kirtland Air Force Base in Albuquerque, one of the most sensitive installations in the American nuclear complex, home to Sandia National Laboratories and the Air Force Weapons Laboratory. He was methodical, technically literate, and serious about documentation. He was exactly the kind of person who notices things.
Beginning around 1979, Bennewitz noticed unusual electromagnetic signals emanating from the direction of Kirtland. He began recording them, systematically, with equipment he built himself. The signals followed patterns he could not match to any known military communication protocol. Separately, he and his family began observing unusual aerial phenomena near the Manzano Mountains and the weapons-storage areas of the base, lights, structured objects, behavior inconsistent with conventional aircraft.
He documented what he saw. He attempted to decode what he heard. And in 1980, he did something that sealed his fate: he wrote to the Air Force and told them what he had found.
"I have been doing a lot of research in these areas and I have some data that I think will be of interest to you."
The Air Force's interest was not what he expected. The base did look into his reports, earnestly, at first. An internal investigation from late 1980, now partially declassified, acknowledged that Bennewitz had recorded genuine signal activity near the installation and that his observations of aerial phenomena were consistent with reports from base personnel during the same period. Several Kirtland security officers filed corroborating reports.
Whatever Bennewitz had picked up, it was real enough to warrant attention from the Air Force Office of Special Investigations. What AFOSI concluded, and what they chose to do about it, would determine the rest of his life.
An intelligence operation has two components: what it protects, and what it produces. AFOSI needed both.
The Air Force's concern about Bennewitz was not that he was wrong. It was more uncomfortable than that. The signals he had intercepted appear, in the fragmentary record that has since become public, to have been related to classified technology testing at Kirtland, possibly associated with directed-energy programs, experimental propulsion concepts, or compartmented surveillance operations run from the base or the adjacent Manzano weapons-storage area. The specific programs have never been fully disclosed.
What is documented, through FOIA releases, congressional testimony, and the on-record statements of multiple intelligence-community veterans, is what AFOSI decided to do next. Rather than simply asking Bennewitz to stop, or, alternatively, bringing him into confidence, they assigned an agent named Richard Doty to manage him.
Richard Doty was an AFOSI counterintelligence specialist stationed at Kirtland AFB. He has since acknowledged, in multiple interviews and in sworn testimony to congressional investigators, that he was assigned to feed Paul Bennewitz fabricated information about extraterrestrial activity as a means of discrediting his legitimate signal intercepts and redirecting his research away from classified programs. He has stated that he acted under orders. The chain of command above him has never been formally identified.
The operation Doty ran was not improvised. It was structured, with a logic that suggests either prior experience with this kind of management, or institutional guidelines for how to handle a civilian who had gotten too close to something real. The approach had several components, deployed over roughly two years, each designed to reinforce the others.
The disinformation fed to Bennewitz was detailed, internally consistent, and designed to be unverifiable. That was the point. Step through the escalation.
The operation achieved its objective. By the time Bennewitz's research had become widely known in the UFO community, it had been so thoroughly contaminated by the fabricated material that separating what he had actually observed from what he had been fed was essentially impossible. Researchers who cited him cited the disinformation alongside the real intercepts, unable to tell which was which. His credibility, and by extension, the credibility of anyone who relied on his work, was effectively destroyed.
This was not an accident. It was the outcome the operation was designed to produce.
"We were trying to deflect him, to send him off in the wrong direction. The idea was to make him look so crazy that nobody would believe him."
The Bennewitz operation was not a rogue action by one agent. The PURSUE corpus is now a partial X-ray of the institutional infrastructure that ran it.
The most important thing to understand about what happened to Paul Bennewitz is not the individual actions of Richard Doty. Doty has acknowledged his role and expressed, in later interviews, something resembling regret. The more significant question is structural: what kind of institution creates the conditions under which this operation is not only possible, but apparently authorized, staffed, documented, and sustained over multiple years?
The PURSUE declassified corpus, hundreds of records from war.gov/ufo spanning FBI, USAF, CIA, DIA, and other agency sources across several decades, does not contain the Bennewitz file. That file, if it exists in preserved form, has not been released. But the corpus does provide something valuable: a wide-angle view of how the institutional apparatus documented, classified, and managed UAP-adjacent information across the same decades in which the Bennewitz operation ran.
Several patterns in the corpus are directly relevant to understanding the operational context Doty was working within.
None of this proves that the Bennewitz operation was formally authorized at the institutional level, the documentation for that specific authorization has not been released. What it does establish is that the operation was not anomalous within its environment. The apparatus that ran it, parallel routing, civilian-witness management protocols, extended classification of sensitive installation programs, is visible in the public record. The Bennewitz operation was not a bug in this system. It was a feature.
Kirtland and the Manzano weapons-storage complex sit at the center of this story; the same nuclear-adjacency pattern recurs across the corpus and is mapped on the Nuclear Proximity Map.
The PURSUE transparency mandate is the first time the institutional apparatus has explicitly moved in the opposite direction. That is worth examining carefully.
In 2025–2026, the executive order directing the declassification and public release of UAP-related government records, and the subsequent publication of the PURSUE corpus at war.gov/ufo, represented something that had not existed before: a formal, public commitment by the executive branch to release the primary-source record rather than manage it.
Whether that commitment is complete, or whether the released corpus represents a carefully curated fraction of the full record, is an open question this investigation takes seriously. But the direction of the action matters. For four decades, the institutional default was containment: classification, redirection, compartmentalization, and, in cases like Bennewitz's, active disinformation. PURSUE is the first formal institutional break from that default.
The break is legible specifically because the Bennewitz case made the old default visible. It is difficult to claim the current transparency mandate as meaningful if you do not also acknowledge what transparency is being offered in contrast to. The context is the apparatus. The reversal is only legible against the background of what was reversed.
The central mechanism of the Bennewitz operation, and of the broader Cold War management of UAP information, was control of the primary source. If you hold the documents, you control what is real. PURSUE removes that control for the released records. The documents now in the public domain cannot be un-released, cannot be reframed without contradiction, and cannot be used as a foundation for disinformation, because they are available to anyone who wants to check. This is structurally different from every prior UAP disclosure arrangement.
This does not mean the PURSUE corpus is complete. The redaction analysis on this site documents significant information removal, whole sections where operational details, location names, and program identifiers have been withheld. The corpus is not a full accounting. It is a first step, meaningful precisely because it is a step in a direction the apparatus did not previously move.
Paul Bennewitz spent the last productive years of his life trying to understand something real using data that had been deliberately corrupted. The PURSUE corpus is, among other things, an attempt to make that kind of corruption impossible going forward, by putting the uncorrupted primary source in the public record before anyone can decide what it should say.
The PURSUE corpus is not an alien artifact. It is the output of a specific institutional apparatus, the same agencies, the same classification architecture, the same interagency coordination structure, that managed UAP information for decades, including through operations like the one run against Paul Bennewitz. Reading the corpus means reading that apparatus. The Bennewitz case is the clearest documented example of what that apparatus was willing to do when it decided the information needed to be controlled. The documents in PURSUE are what it looked like when the apparatus decided, or was directed, to let the information go instead.