The question isn't "do hidden things exist." That's settled — they do, and we have receipts. The question is which categories of hidden things tend to be real, which tend to be embroidered, and which tend to be projection. A serious investigator works the priors, not the vibes.
Two failure modes corrupt almost every conversation about "shadow programs." The credulous mode treats every leaked PowerPoint, every anonymous airman, every grainy IR clip as load-bearing — building skyscrapers of inference on foundations of single-source testimony. The dismissive mode runs the opposite play: because some claims are absurd, all claims must be. Both modes are intellectual laziness. Neither survives contact with the documentary record.
What follows is an honest attempt at a third path. We treat each claim like a GYI investigation thread — same A/B/C/D tiering, same evidentiary standards. If a claim earns a Tier A on the public record, we say so. If it earns a Tier D, we say so. If a claim is genuinely unresolved, we name what would resolve it.
The tiering system is the same one we used on the Red River Turbines, Choctaw Global, CHIPS, EDTX, and Cannell threads. It's not optimized for storytelling. It's optimized for not embarrassing ourselves.
| Tier | Standard | Example phrasing |
|---|---|---|
| A | Confirmed by primary-source documentation: congressional record, declassified IG report, court filing, sworn testimony plus corroboration, agency-released document. | "This happened. Here is the paper." |
| B | Strong circumstantial: multiple independent sources, journalistic investigations with named sourcing, whistleblower testimony with IG corroboration but no released artifact. | "Something is here. Reasonable observers conclude X." |
| C | Suggestive but thin: single-source claim, plausibility argument from structure, pattern recognition without artifact. | "It's possible and worth a FOIA, but don't bet on it." |
| D | Failed the test: contradicted by primary documents, traced to fabrication, or unfalsifiable in principle. | "Debunked or unfalsifiable. Move on." |
The rules of engagement: extraordinary claims require extraordinary evidence, but ordinary claims require ordinary evidence — not zero evidence. The fact that some category of program exists in general doesn't promote a specific claim within that category. Conversely, the fact that some specific claim is unproven doesn't demote the category.
Before assessing what might be hidden now, we need to know the base rate: how often does the US government actually run major secret programs that later become undeniable? The answer is: often. Often enough that "this could be hidden" is not a low-prior claim. Here are the canonical Tier A confirmations — all proven, all once denied, all uncovered through some combination of leaks, congressional investigation, court action, or whistleblower disclosure.
This list is partial. It excludes well-documented but less famous cases: Operation Mockingbird (CIA-press relationships), the Tuskegee syphilis study (1932-1972, ended only after AP exposure), Operation Sea Spray (1950 SF Bay biological-warfare simulation, declassified 1976), Project SHAMROCK (1945-1975 NSA cable interception), Project Iceworm (nuclear under Greenland ice, declassified 1996), the Family Jewels (679-page CIA internal misconduct catalog, released 2007). And it stops at fully-declassified items — it doesn't include programs that are partially acknowledged, like the National Reconnaissance Office's existence (denied 1961-1992).
From 1947 to present, every decade has produced at least one major program that was officially denied while operating, then later confirmed by primary-source documentation. The base-rate prior on "the US government runs significant programs the public doesn't know about" is therefore not a fringe claim. It's the historical default.
What this doesn't tell us: which currently-suspected programs are real. A high base rate for the category doesn't promote any specific claim within the category.
To assess what could plausibly be hidden, you have to know the actual architecture of US government secrecy — not the movie version. It is more layered, more legal, and more compartmented than most people realize. This is not theory; it is DoD policy, published in DoD Manual 5205.07 and Executive Order 13526.
The "Gang of Eight" briefing structure is the most consequential governance fact in this entire question. By law and longstanding practice, certain SAPs are briefed only to eight lawmakers — the Speaker, Minority Leader, Majority and Minority Leaders of both chambers, plus the chairs and ranking members of the two intelligence committees. These individuals cannot share what they learn with their staff, with their full committees, or with each other in writing. They cannot read it twice from the same document. They cannot take notes.
This means a program can be legally classified at a level where fewer than ten elected officials have any visibility into it. Whether you call this "necessary security" or "absent oversight" depends on the program — but the structure is real, codified, and Tier A confirmed. Any conversation about "what could be hidden" that doesn't engage with the Gang of Eight architecture is missing the point.
When David Grusch testified that a crash-retrieval program exists at the Waived USAP level, he was making a structural claim that maps onto an existing, legally-real architecture. That doesn't make the claim true. But it does mean the claim is at minimum not architecturally impossible — the secrecy infrastructure capable of housing such a program is documented and operating. The question isn't whether the box exists. The question is whether anything is in it.
The "black budget" is a real, countable thing — not folklore. Since 2007 the Director of National Intelligence has been required by statute to publish the topline National Intelligence Program budget. The Military Intelligence Program topline is also published. What you cannot get publicly is the breakdown by agency or program below those toplines.
The structural point: roughly the size of Hungary's GDP is being spent annually on activities that not even the appropriating committees see in full detail. Of that, an unknown but non-trivial fraction flows to Waived USAPs — programs that cannot be discussed in any congressional session that isn't a Gang of Eight read-in. This is the answer to "could anything significant be hidden?" The answer is yes, mechanically and financially. The unanswered question is what specifically is being concealed at that level.
From December 2017 onward, the US government has done something it had not done in the previous half-century: openly acknowledge that its own pilots and sensors record objects it cannot identify, take the question seriously enough to fund an office, and convene congressional hearings under oath. Below is the timeline of only the events confirmed by primary-source documentation. We're excluding everything that depends on a single anonymous source.
Here is the assessment for the major specific claims, tiered the same way we tiered Red River Turbines, Choctaw Global, and the rest. The grading is honest: where the evidence is there, we say so; where it isn't, we say so; where it's split, we name the split.
| Tier | Claim | What's actually documented |
|---|---|---|
| A | The US government runs Special Access Programs that are denied to non-cleared personnel and Waived USAPs briefed only to the Gang of Eight. | DoD Manual 5205.07; EO 13526; routine GAO and CRS analysis. Architecturally settled. |
| A | The combined NIP+MIP intelligence topline is ~$106B/yr (FY25) and the program-level breakdown below the topline is classified. | ODNI annual public release; statutory requirement since 2007 (PL 110-53). |
| A | US Navy aviators and ship sensors recorded objects that the Navy itself could not identify and the Pentagon publicly released those recordings. | defense.gov, April 27, 2020 official release of FLIR1, GIMBAL, GOFAST videos. |
| A | AATIP (2007-2012) and its successors UAPTF and AARO are real, congressionally-funded offices. | FY23 NDAA, Pub.L. 117-263 § 1683; AARO charter; congressional testimony by Susan Gough. |
| B | Multiple credible whistleblowers, after ICIG vetting found their complaints "urgent and credible," have testified under oath that crash-retrieval programs exist. | Grusch July 2023 testimony; ICIG correspondence; multiple corroborating military sources unnamed in public testimony but referenced by Grusch as having submitted protected disclosures. No released artifact. |
| B | Some fraction of UAP cases involve sensor signatures consistent with vehicles outperforming known US, Russian, or Chinese aircraft. | ODNI 2021 Preliminary Assessment (18 of 144 showed unusual movement); 2004 Nimitz USS Princeton radar tracks corroborated by multiple operators; FLIR1 anomalous trajectory. Alternative explanations (parallax, IR artifact) remain live for some clips. |
| B | Adversary capabilities (Russian/Chinese hypersonics, surveillance drones) account for a non-trivial fraction of "UAP" incidents, including some originally reported as anomalous. | DoD off-record briefings; pattern of Chinese balloon overflights (Feb 2023); pattern of Russian drone overflights of US bases reported by AARO. |
| C | Crash retrieval program contains non-human biologics or vehicles of non-human origin. | Grusch sworn testimony; no released artifact; AARO Vol 1 explicitly concluded the opposite based on its records review. Genuine evidentiary standoff with no public-domain tie-breaker. |
| C | Defense contractors hold reverse-engineered technology outside government oversight. | Asserted by Grusch citing named (but congressionally-only-disclosed) corporations; named by Eric Davis to AATIP per leaked Wilson-Davis memo (disputed authenticity). Pattern-consistent with how SAPs are structured but no document trail. |
| C | The 2004 Nimitz "Tic Tac" represents a vehicle whose physics is not explained by any known US or adversary capability. | Multiple-operator corroboration; radar + IR data; corroborated by Fravor, Slaight, Dietrich, Day. Alternative explanations (sensor anomaly, classified US program) remain logically possible but the multi-platform, multi-operator nature is unusual. |
| D | The MJ-12 documents are authentic 1947-1980s briefings on crash retrieval. | FBI investigated and concluded "completely bogus"; document format inconsistencies; signatures don't match known exemplars. Forgeries. |
| D | Bob Lazar worked at S-4 reverse-engineering alien craft as a physicist. | Lazar's claimed degrees (MIT, Caltech) returned no records on multiple independent inquiries. EG&G employment claim never corroborated. Specific physics claims (element 115 antigravity) contradicted by later discovery of element 115's actual properties. |
| D | The 1947 Roswell incident involved an extraterrestrial craft. | USAF's 1994 and 1997 reports identified the debris as Project Mogul (classified high-altitude balloon array for Soviet nuclear-test detection). The contemporaneous secrecy is real; the explanation has held up under 30 years of skeptical scrutiny. |
| D | The Phoenix Lights (March 13, 1997) were a single craft of unknown origin. | Two distinct events conflated: (1) high-altitude formation of A-10 aircraft on training mission; (2) flare drop by Maryland ANG over Barry Goldwater Range. Both events have multiple independent confirmations including base records. |
The architecture is Tier A. The disclosure-era government posture change is Tier A. The whistleblower complaint mechanism producing "urgent and credible" findings is Tier B. The specific claim that any of this points to non-human intelligence is Tier C — not debunked, but not confirmed by any artifact in the public domain. The popular 20th-century artifacts (Roswell, Lazar, MJ-12, Phoenix Lights) are Tier D and should not be cited.
The "confirmed hidden" column is dominated by government misbehavior, surveillance, and weapons R&D. The "debunked" column is dominated by esoteric / extraterrestrial / metaphysical claims with no documentary backing. This pattern holds across 70 years. When something hidden becomes public, it is usually a classified program, an illegal surveillance operation, an unethical experiment, a covert action, or a weapons capability — not crashed UFOs, not reptilians, not occult cabals. The exceptions are vanishingly rare in the historical record.
This is not proof that the UAP question is closed. It is a calibration data point. Any honest investigator should weight prior probabilities accordingly.
Across roughly 80 confirmed major secret programs from 1945 to present (collated from Church Committee, Pike Committee, Rockefeller Commission, Snowden disclosures, Senate Select Committee on Intelligence reports, Family Jewels release, court orders, declassification reviews, and ProPublica/Intercept/NYT/Post investigations), the distribution by type breaks roughly as follows:
The structural takeaway: over 80 years, when "what was hidden" got pulled into the light, it has almost always been something the government considered tactically necessary to deny but didn't involve exotic-physics or extraterrestrial elements. The categories that consistently produce disclosures are surveillance, covert paramilitary, advanced-but-not-impossible weapons systems, and unethical research on human subjects.
This doesn't mean an exotic-category disclosure is impossible — the base rate of a category producing its first major disclosure is non-zero. It means that the prior probability for the next major disclosure being in the exotic category is substantially lower than for the conventional categories. If you're betting on what next year's "shocking declassification" is, the smart money is on another surveillance program or another covert-action operation. The smart money is not on bodies on ice at Wright-Patterson.
Here, finally, is the honest verdict on the question you asked.
If a thoughtful reader takes nothing else from this brief: secrecy is real, large, and bigger than most casual observers grasp. The categorical history of what gets hidden strongly favors surveillance, weapons, and covert action over exotic phenomena. The UAP question is the most credible "exotic" thread in a generation, but it sits at Tier C on its specific claims, not Tier A.
You asked what we'd find if we applied our talents honestly. The answer is: more boring than the conspiracy and stranger than the official denial. The shadow programs are real. They are probably not what most people think they are. And the actual hidden things, when they surface, tend to be the kind of governmental misbehavior that makes intelligence-committee reports embarrassing, not the kind that rewrites the physics textbook.
That said: the Tic Tac case is genuinely unexplained, the Grusch testimony is genuinely on the record, and the disclosure-era infrastructure (AARO, congressional hearings, statutory records review) is genuinely something that did not exist before 2017. Reasonable observers can place those facts inside either a "this is leading somewhere" or "this is a contained civic process" interpretation. We are not telling you which. We are telling you not to take any source — including this one — as the last word.
Nothing in this brief should be taken on faith. Everything cited above can be checked against primary sources. The following resources are where to start.
| Resource | What's there |
|---|---|
| Federation of American Scientists — Project on Government Secrecy | Steven Aftergood's archive of classification policy, budget releases, IG reports. fas.org/issues/government-secrecy |
| National Security Archive (GWU) | Largest non-governmental repository of declassified US national security documents. Searchable. nsarchive.gwu.edu |
| CIA CREST Database | 13M+ pages of declassified CIA documents searchable online. cia.gov/readingroom |
| AARO official site | Historical Record Report Vol 1, public submissions, FAQ. Government voice; treat skeptically but read it. aaro.mil |
| ODNI public releases | Annual statistical transparency reports, Section 702 reports, NIP toplines. dni.gov/files/documents |
| Church Committee Final Report | 1976 Senate Select Committee on Intelligence Activities — the foundational US declassification of IC misconduct. Still indispensable. intelligence.senate.gov/resources/intelligence-related-commissions |
| SSCI Detention & Interrogation Study | Declassified summary, December 2014. The definitive document on post-9/11 CIA black sites. intelligence.senate.gov/study2014 |
| FOIA.gov + MuckRock | File your own requests; MuckRock aggregates other people's results. Free and powerful. |
If you cannot find a claim in a primary source (court filing, IG report, congressional record, declassified document, named-source contemporaneous reporting), and it isn't directly attributed to a person who would testify under penalty of perjury, do not adopt it as a belief. You can hold it as a hypothesis. You cannot reason from it as fact. This rule, applied consistently, will save you from 90% of the noise and let you concentrate on the 10% that's actually load-bearing.