CIA Project Stargate: The Complete Declassified Picture

In 2017, the CIA published roughly 12 million pages of declassified documents in a searchable online archive. Within them, about 12,000 documents covered a single program: Stargate. Until that release, most public information about U.S. government remote-viewing research had been filtered through journalistic accounts, memoirs, and partial declassifications. The 2017 release made the primary record available for direct examination.

What it shows is more interesting — and more nuanced — than either skeptical or enthusiast accounts had previously suggested.

What was Stargate, technically?

"Stargate" is the umbrella name for a 23-year (1972-1995) U.S. government effort to investigate and operationalize remote viewing — the alleged ability to perceive distant or hidden targets without sensory access.

The program ran under several names across its history:

• SCANATE (1972-1976) — initial CIA-funded research at Stanford Research Institute (SRI)
• GONDOLA WISH / GRILL FLAME (1977-1979) — Army-funded operational testing
• CENTER LANE (1983-1986) — DIA-managed expansion
• SUN STREAK (1986-1991) — DIA continuation
• STAR GATE (1991-1995) — final consolidated phase

The research arm was at SRI International (Menlo Park, CA), led by physicists Hal Puthoff and Russell Targ. The operational arm was at Fort Meade, Maryland, with a unit of military "viewers" including Joe McMoneagle, Lyn Buchanan, Mel Riley, and others.

[CIA Reading Room. STAR GATE archive.]

What did the laboratory research find?

Two decades of laboratory research at SRI (and later at SAIC) tested remote viewing under increasingly controlled conditions. Key findings:

Coordinate Remote Viewing (CRV) protocol: Developed by Ingo Swann + Hal Puthoff. Viewers given only geographic coordinates were asked to describe the location. Independent judges then matched descriptions to photographs of the actual locations vs control locations.

Results across the SRI program:

• Hit rates significantly above chance
• Effect sizes of ~Cohen's d = 0.4-0.5 in pooled data
• Some viewers consistently outperformed others
• The "decline effect" — performance dropping over years of testing — was observed

Specific operational successes:

• Pat Price's description of the Soviet weapons facility at Semipalatinsk (1974) — a CIA-validated description that included specific gantry crane configurations later confirmed by satellite imagery. This is the most-cited operational success.
• Joe McMoneagle's description of a Soviet submarine class under construction at Severodvinsk (1979) — described features later identified as the Typhoon class, two years before official Western intelligence confirmation.
• Description of a downed Soviet Tu-22 in Africa (1979) — President Carter mentioned this publicly years later as an instance of "psychic" intelligence locating a downed plane.
• Iran hostage crisis (1980) — viewers were tasked with locating American hostages; results were mixed but included some confirmed-accurate details.

[Targ R. (2012). The Reality of ESP: A Physicist's Proof of Psychic Abilities. Quest Books.]

The 1995 AIR review and termination

In 1995, Congress requested the American Institutes for Research (AIR) to evaluate the Stargate program before continued funding decisions. AIR commissioned two reviewers:

• Jessica Utts (statistician, UC Davis) — reviewed the laboratory data
• Ray Hyman (psychologist, U. Oregon) — reviewed both lab and operational data

Utts's conclusion: The statistical evidence for remote viewing was "beyond reasonable doubt" — effect sizes were small but consistent and replicated. She wrote: "Using the standards applied to any other area of science, it is concluded that psychic functioning has been well established."

Hyman's conclusion: While Hyman acknowledged the statistical evidence was non-trivial, he argued that:

• Methodological flaws couldn't be ruled out as explanations
• Operational utility was insufficient for intelligence purposes
• Effects, even if real, were not consistent enough for actionable intelligence

The CIA's decision: Terminate the program. The official reasoning cited the operational utility judgment, not the absence of statistically significant effects. The CIA acknowledged the data showed something above chance — they simply concluded it wasn't useful enough operationally.

This is one of the most underreported aspects of the Stargate story. The official termination did NOT conclude "remote viewing isn't real." It concluded "remote viewing isn't operationally actionable enough for intelligence work."

[American Institutes for Research. (1995). An Evaluation of Remote Viewing: Research and Applications.]

What did the 2017 release actually show?

The 12,000+ documents released in 2017 added important detail to the publicly known story:

Operational targeting was extensive. Documents show remote viewing was used (or attempted) for: Soviet underground facilities, Libyan chemical weapons, Iranian hostages (1980), Beirut hostage situations (1980s), Nicaraguan Sandinista activities, Aldrich Ames and other counterintelligence targets, narcotics trafficker locations.

The unit was operationally embedded. Fort Meade viewers were not just doing research — they were a working intelligence unit producing requested products for operational customers (DIA, Army, sometimes CIA).

Quality control existed. Multiple internal evaluation processes existed; not every viewer was retained; performance was tracked and reviewed.

Skepticism was present internally. Many internal documents show analysts and managers debating the program's value. The decision to terminate was not externally imposed — internal skepticism was consistent across the program's history.

Costs were modest. Total program cost over 23 years: $20 million ($50 million in 2026 dollars). For comparison: a single F-35 unit cost more in 2024.

[CIA. (2017). STAR GATE program documents archive.]

What are the key viewers' accounts?

Several Stargate viewers wrote detailed memoirs after declassification:

Joe McMoneagle — Remote Viewing Secrets (2000), Mind Trek (1993). McMoneagle is widely considered the unit's most reliable viewer, with documented operational successes (the Severodvinsk submarine, the Iran hostage descriptions).

Lyn Buchanan — The Seventh Sense (2003). Detailed account of CRV training and unit operations from a viewer's perspective.

Paul Smith — Reading the Enemy's Mind (2005). Most thorough account of unit history from inside Fort Meade.

Russell Targ — The Reality of ESP (2012), Limitless Mind (2004). Co-founder of the SRI program, focuses on laboratory science.

Hal Puthoff — has not written a comprehensive memoir but has given numerous interviews and presentations covering the SRI work.

These accounts agree on broad strokes but differ on details. They form a partial corroborating picture, though they're memoirs by participants — not independent verification.

What are the strongest skeptical responses?

1. Selection bias in success stories. Viewers produced thousands of session reports. Critics argue the cited "successes" are cherry-picked from a much larger corpus that included many misses. Without seeing the full session inventory, the hit-to-miss ratio is unclear.

The 2017 release does include operational reports that were judged by reviewers to be unsuccessful. The picture is more mixed than the success stories alone suggest. But the laboratory data (where complete sessions were tracked) does show statistically significant above-chance performance.

2. Insufficient blinding in some sessions. Some operational sessions had compromised blinding (viewers received briefings with target context). The laboratory sessions were better controlled.

3. The 1995 termination is the only fact that matters. Skeptics argue that whatever the data showed, the U.S. government — with the strongest possible incentive to continue if the program worked — chose to terminate. This is meaningful evidence about operational utility, even if not about phenomenon reality.

Counter-argument: bureaucratic decisions don't always track scientific evidence; many programs survive that shouldn't and many die that should have continued. Termination is a data point, not the verdict.

What does Stargate establish?

Well-established:

• A serious 23-year U.S. government program existed
• Laboratory research produced statistically significant above-chance effects
• Some operational targeting produced confirmed-accurate descriptions
• The program was reviewed and judged operationally insufficient in 1995
• The 2017 declassification provides extensive primary-source documentation

Not established:

• That remote viewing is reliable enough for operational intelligence
• That every cited "operational success" was confirmed independently
• That the underlying mechanism is understood
• That continued investment would have produced better results

Open:

• Why the U.S. terminated while related programs (alleged) in other countries did not
• Whether subsequent classified programs continued the work
• The current operational status

How does this connect to the broader pillar?

Stargate sits within the government psi programs pillar as the most thoroughly declassified and best-documented example. Related work covered in:

• Soviet psi research during the Cold War — parallel program in the USSR
• MK-ULTRA consciousness experiments — earlier CIA program targeting consciousness
• Gateway Process CIA document explained — adjacent CIA-funded consciousness training program

For the broader pillar overview, see Government Psi Programs.

Sources

• CIA Reading Room. STAR GATE archive (declassified 2017).
• American Institutes for Research. (1995). An Evaluation of Remote Viewing: Research and Applications.
• Targ R. (2012). The Reality of ESP: A Physicist's Proof of Psychic Abilities. Quest Books.
• McMoneagle JW. (2000). Remote Viewing Secrets: A Handbook. Hampton Roads.
• Smith PH. (2005). Reading the Enemy's Mind: Inside Star Gate. Forge Books.
• Utts J. (1996). An Assessment of the Evidence for Psychic Functioning. Journal of Scientific Exploration, 10(1), 3-30.