Adult Psychiatry· 18 min read

MKUltra: The CIA’s Psychedelic Experiments and Informed Consent

Between 1953 and the early 1970s, the Central Intelligence Agency ran experiments that gave powerful drugs to people who never agreed to take them. The program was called MKUltra. Its subjects included prisoners, psychiatric patients, people who used drugs, the men lured into its safehouses, and, in at least one documented subproject, children. Most of them never knew what had been done to them.

I want to be careful with this topic, because it attracts a lot of invention. This article stays with the documented record: the Church Committee, the 1977 Senate hearing, the court opinions, and the roughly 16,000 pages of files that survived the CIA’s own shredder. Where the record is disputed, I say so. Where it has been exaggerated, I leave the exaggeration out.

University labs were studying psilocybin and LSD in the open in those same years. I tell that story in the history of psilocybin. The CIA was studying the same drugs on unwitting Americans, as tools for interrogation and control.

Two things make this a story a psychiatrist should tell. One is informed consent, the principle that a person has to understand and agree to what’s done to them. MKUltra ran on its absence. The other is accountability. Almost no one was held to account, and when the victims went to court, they lost.

Frank Olson: A Death the Government Hid for 22 Years

The most documented casualty of MKUltra was one of the government’s own. Frank Olson was an Army biological-warfare scientist at Fort Detrick who worked closely with the CIA. In November 1953, Gottlieb’s team gave him LSD without telling him.

It happened at a working retreat at Deep Creek Lake, Maryland. On the evening of November 19, Gottlieb’s deputy spiked a bottle of Cointreau the group was sharing, and about 20 minutes later told the men they had just taken LSD. Olson grew agitated and paranoid over the following days.

Nine days later, at around two in the morning on November 28, Olson went out a window of the Statler Hotel in New York City and fell to his death. He had been taken there, ostensibly, to see a CIA-connected doctor about his state of mind. The official finding was suicide. His family was told nothing about the LSD and lived with that account for 22 years.

The truth surfaced by accident. In 1975, the Rockefeller Commission’s report on CIA abuses described a scientist who had been dosed with LSD and died soon after, and the Olsons recognized their father and husband in the anonymized account. President Ford apologized to the family in the Oval Office, and in 1976 Congress passed a private law paying them $750,000 in exchange for dropping their legal claims.

The forensic dispute centers on the body itself. In 1994 the family had it exhumed, and James Starrs, the scientist who re-examined it, found a head injury he called “rankly and starkly suggestive of homicide.” That was his professional opinion. It was never proven, and no one was charged.

When the sons sued the government again in 2012, a federal district court dismissed the case in 2013 as filed too late and barred by the 1976 settlement. The judge never reached the question of whether Olson was murdered.

The manner of Frank Olson’s death remains unresolved. Whatever killed him, the consent violation is settled: his own colleagues gave him a powerful hallucinogen without warning, and nine days later he was dead.

What MKUltra Actually Was

MKUltra started as a memo. On April 3, 1953, Richard Helms, then a senior officer in the CIA’s clandestine service, proposed a covert program to study chemical and biological ways of controlling human behavior. Director Allen Dulles approved it 10 days later, on April 13, 1953.

The man who ran it was a chemist named Sidney Gottlieb, chief of the Chemical Division of the CIA’s Technical Services Staff. The “MK” in the name marked it as a Technical Services project. Gottlieb would direct the work for most of the next 20 years.

At the August 3, 1977 Senate hearing that put MKUltra on the public record, investigators counted 149 subprojects across 80 institutions: 44 colleges and universities, 15 research foundations or pharmaceutical companies, 12 hospitals and clinics, and 3 prisons.

The money moved through cutouts, so the people taking it often didn’t know the CIA was the source. The main conduit was the Society for the Investigation of Human Ecology, joined by the Geschickter Fund for Medical Research and the Josiah Macy, Jr. Foundation.

One foundation keeps getting pulled into this story, and the record doesn’t put it there. The Foundations’ Fund for Research in Psychiatry was signed into existence on February 14, 1953, two months before Dulles approved MKUltra. It was private philanthropy. A wealthy former psychiatric patient named Charles B.G. Murphy built it, with Yale as trustee.

The field was so starved for money then that the psychoanalyst Lawrence Kubie had titled a 1952 paper in Science “Research in Psychiatry Is Starving to Death.” Murphy’s fund was the open answer to that problem. The CIA’s fake foundations were the covert one.

A sealed 1950s-style glass ampoule with a blank inventory tag resting in a ceramic dish, representing the LSD the CIA administered without consent

That fund’s money is accounted for. Murphy gave about $12 million of his own, and the Ford Foundation added $4.5 million. Over 29 years it spent roughly $18.5 million on researchers like Erik Erikson, Anna Freud, and David Hubel.

Hubel’s $8,500 grant from the fund in 1957 kept him in the visual-cortex work that later won a Nobel Prize. The American Journal of Psychiatry’s history of the fund credits it with helping to launch the modern era of psychiatric research.

I’ve checked the claim that this fund came out of MKUltra. The primary record says otherwise: the 1977 Senate hearing, the Church Committee reports, and the Marks files all name the real cutouts, and none of them mentions this fund. The likely source of the confusion is that the genuine fronts had near-identical names.

LSD was the program’s main drug, and 10 of the first 50 subprojects involved it. But the program tested a whole pharmacy, including psilocybin, mescaline, and several depressants, alongside hypnosis and electroshock.

The idea behind all of it was chemical control of the mind. The people the CIA tested that idea on were chosen because no one would come looking for them.

Operation Midnight Climax

Some of the testing was done on people who had walked in off the street. To run it, Gottlieb recruited George Hunter White, a federal narcotics agent lent to the CIA by the head of the Federal Bureau of Narcotics, Harry Anslinger.

White set up safehouses, first in New York in 1953, then in San Francisco and Marin County starting in 1955. Paid sex workers brought men back to the apartments and slipped them LSD without their knowledge, and White watched through a two-way mirror, reportedly with a pitcher of martinis.

The men were never told what they’d been given, and no one followed up to see what it did to them. The operation had a name inside the Agency: Operation Midnight Climax.

The CIA’s inspector general, John Earman, questioned the whole arrangement in a 1963 report. The safehouses stayed open anyway, San Francisco until 1965 and New York until 1966. White retired that same year.

When it was over, White wrote Gottlieb a letter that reads like a confession. “Where else could a red-blooded American boy lie, kill, cheat, steal, rape, and pillage with the sanction and blessing of the All-Highest?” That was the man the CIA trusted to run human experiments in an apartment with a two-way mirror.

The Mushroom the CIA Chased

The CIA went after psilocybin too, and it went after it early. In 1956, the Agency wanted mushroom samples from Mexico. It put $2,000, disguised as a grant from the Geschickter Fund, toward an expedition led by the banker and amateur mycologist R. Gordon Wasson, and it planted its own chemist on the trip to bring specimens back. That was MKUltra Subproject 58.

Those were the same mushrooms from which Albert Hofmann isolated and synthesized psilocybin in 1958. I trace that discovery in the history of psilocybin. Hofmann worked in the open at Sandoz. The CIA wanted the same molecule for interrogation, to find out whether it could break a person’s will.

Psilocybin itself is a serotonergic drug that acts mainly through the 5-HT2A receptor, the same mechanism behind the antidepressant effects now under study. The CIA was interested in the opposite: whatever could disorient or coerce.

Two dried mushroom specimens mounted on an aged herbarium sheet, representing the 1956 expedition samples the CIA funded through Subproject 58

The subjects included children. Reading the surviving documents, the historian Stephen Kinzer describes a subproject in which 12 pre-pubescent boys were given psilocybin, and a parallel experiment in which children with schizophrenia were given LSD daily for weeks. This comes from Kinzer’s reading of the MKUltra files, and no court ever ruled on it. But the files are real.

The Shredder, and the Files That Survived

In 1973, on his way out of the CIA, Richard Helms ordered the MKUltra files destroyed. On January 30, seven boxes were shredded at a records center in Warrenton, Virginia. Gottlieb reportedly drove out to make sure it happened, and had the files in his office safe destroyed too.

A separate cache of financial records had been filed in the wrong place and escaped the shredder. When a former State Department officer named John Marks filed a Freedom of Information Act request in 1977, under a new CIA director, Stansfield Turner, an archivist found those boxes.

That cache, about 16,000 pages, is most of what anyone knows about MKUltra. Add the Church Committee’s 1975 and 1976 investigations and the August 1977 Senate hearing, and you have nearly everything there is.

Everything credible written about the program traces back to those sources. Some counts of the surviving pages run higher than 16,000, but that’s the figure they support.

The Courthouse Door: United States v. Stanley

The people these programs harmed had almost no way to make the government answer for it. The clearest proof of that came from a case the Army caused, in a chemical-warfare program that ran alongside the CIA’s. It went all the way to the Supreme Court, and the plaintiff still lost.

Master Sergeant James Stanley volunteered in February 1958 for what he was told was a test of protective equipment at the Army’s Edgewood Arsenal in Maryland. He was given LSD, secretly, about 4 times. He survived it. But he came out with hallucinations, memory loss, and episodes of violence that ended his marriage, and the Army discharged him in 1969.

When Stanley sued, the Supreme Court ruled against him. In United States v. Stanley, decided 5 to 4 in 1987, the Court held that a soldier can’t sue the government for injuries that happen “incident to service,” even when the injury is a secret drugging that violates the Constitution.

The legal shield is the Feres doctrine, and it left Stanley with no case at all. Congress later pushed for compensation, reported at around $625,000. The ruling stood.

Put Stanley next to Frank Olson and the pattern is hard to miss. Olson’s sons got the same answer in 2013. They were two of the most documented victims of Cold War drug experiments, and neither could get a court to hear the merits.

How Informed Consent Became Law

The reform came through regulation instead, and it changed how every human experiment in the country is run.

Tuskegee was exposed in 1972: the U.S. Public Health Service study that watched Black men with syphilis go untreated for 40 years. MKUltra became public in the same few years.

Tuskegee did more than anything else to drive the reforms that followed.

Congress passed the National Research Act in 1974, and it created the commission that produced the Belmont Report in 1978. Belmont set out the three principles that still govern research on people: respect for persons, the principle informed consent comes from, along with beneficence and justice.

Those principles became federal regulation. By 1991, more than a dozen federal agencies had adopted the Common Rule, which requires an institutional review board and documented informed consent for research on human subjects.

MKUltra breaks all three principles at once. It used deception where consent was required. It harmed people it was supposed to protect, and it picked prisoners, patients, and children because the system treated them as expendable.

The modern rules read almost like a point-by-point answer to what the CIA did. Every study in the current psychedelic therapy revival runs inside that framework.

An uncapped fountain pen resting on an unsigned consent document on warm linen, representing the informed consent framework built after the CIA experiments

The Modern Test: The MDMA Trials

Safeguards only protect people when someone actually uses them.

Between 2014 and 2015, a Phase 2 study of MDMA-assisted psychotherapy for PTSD ran at a site in Vancouver, sponsored by a nonprofit called MAPS, the Multidisciplinary Association for Psychedelic Studies. One participant, Meaghan Buisson, has spoken publicly under her own name about what happened to her in it.

Her therapists were a married couple, Richard Yensen and Dr. Donna Dryer. Yensen held a PhD but wasn’t a licensed psychologist, and because British Columbia didn’t regulate counselors, no professional college had authority over him. Dryer was a licensed psychiatrist.

MAPS ran cameras in the room, as its protocol required. The footage, reviewed by journalists in 2022 and not disputed by MAPS, shows the therapists blindfolding, cuddling, spooning, and physically pinning down a visibly distressed patient during her dosing sessions.

The treatment manual allowed limited nurturing touch, and MAPS itself later found the therapists had substantially deviated from it. What the cameras caught went well past the manual.

After the dosing sessions ended, Yensen entered into a sexual relationship with Buisson while she was still enrolled in the study. He admits the sexual contact. Buisson has alleged it was sexual assault, and Yensen disputes that, saying it was consensual.

Police recommended charges. Crown prosecutors declined to approve them, and Yensen has never been criminally charged. A civil suit Buisson filed in 2018 was settled out of court.

The safeguards that were supposed to catch this all failed in the same direction. MAPS recorded the sessions on its own cameras, and it has acknowledged that no one on its staff watched the footage until 2021, six years after it was filmed.

The trial’s monitoring never flagged the boundary violations. Dryer, the co-therapist who knew about the sexual relationship, didn’t report it, and she resigned her medical license in 2023 after admitting unprofessional conduct.

Buisson and the journalists who broke the story have accused MAPS of downplaying and covering up what happened. MAPS says it reported the misconduct to regulators once it learned of it in 2018, and it calls many of the broader criticisms inaccurate. What the record proves is narrower than a cover-up and still damning: oversight that existed on paper and didn’t function when it mattered.

By the time the treatment reached the FDA, the sponsor had reorganized and renamed itself Lykos Therapeutics. In June 2024, an FDA advisory committee voted 9 to 2 that the trials hadn’t shown MDMA-assisted therapy works, and 10 to 1 that its benefits didn’t outweigh its risks. In August 2024, the FDA declined to approve it and asked for another Phase 3 trial.

The FDA’s stated reasons were about the science: inadequate reporting of adverse events, with inspectors finding unreported events at more than one site, along with problems in how the trials were blinded. The misconduct case was part of the scrutiny around the application. How much it weighed in the decision isn’t clear, and I won’t claim it was the reason.

In August 2024, the journal Psychopharmacology retracted three MDMA-therapy papers built on the Phase 2 data, citing unethical conduct at the Vancouver site that the authors knew about and didn’t disclose. The new trial the FDA asked for hadn’t begun as of mid-2026, and the sponsor has renamed itself again, to Resilient Pharmaceuticals.

That consent problem is not only history. It resurfaced in the modern trials, where a drug that lowers a patient’s defenses made conduct harder to police. See how MDMA-assisted therapy works.

Psychedelic research deserves to continue. The clinical signal for psilocybin in treatment-resistant depression is real, and the case for careful, monitored trials is strong. The MDMA case is a warning about execution, and it lands on the same principle MKUltra did.

If you’re dealing with PTSD or depression right now, the established, FDA-approved treatments don’t require waiting on any of this. I see patients across North Carolina by telehealth, and you can book a consultation to talk through what fits.

A person in a psilocybin trial today signs pages of consent, in a monitored room, under FDA and DEA oversight and an ethics board’s review. That whole apparatus exists because people like Frank Olson didn’t get any of it. The current debate over psychedelic policy is about whether we remember why those rules are there.

A 2026 Postscript: Congress Holds an MKUltra Hearing

On June 30, 2026, the House Oversight Committee’s Task Force on the Declassification of Federal Secrets held a hearing titled “Mind Control and Accountability: Uncovering the Truth of the CIA’s MKULTRA Experiments.” Representative Anna Paulina Luna of Florida chairs the task force. The witnesses were Stephen Kinzer, whose account of the surviving files I cite in this article, and the investigative reporter Tom O’Neill.

Luna called the 1973 shredding “criminal destruction of federal records,” and the task force demanded full declassification. She also says the CIA has started declassifying newly located MKUltra files.

The National Security Archive, which holds the Marks files, published its own briefing the day before the hearing. It told the task force to focus on the real gaps in the record: the seven boxes Helms destroyed and the still-redacted files from MKUltra’s precursor programs, Bluebird and Artichoke. All five documents the archive posted with the briefing had CIA redactions.

MKULTRA FAQ

Was MKUltra real, or is it a conspiracy theory?

Real and documented. The CIA confirmed it under oath. The Church Committee investigated it in 1975 and 1976, a Senate hearing in 1977 laid out its scope, and about 16,000 pages of the program’s records survive. The uncertainty is at the edges, like the exact number of deaths or the manner of Olson’s death.

Did the CIA really experiment with psilocybin?

Yes. Through MKUltra Subproject 58, the CIA funded R. Gordon Wasson’s 1956 mushroom expedition to Mexico and planted a chemist to bring back samples. The surviving record, as read by the historian Stephen Kinzer, also describes subjects given psilocybin, including children. LSD got more attention inside the program, but psilocybin was part of it from early on.

Was the Foundations’ Fund for Research in Psychiatry a CIA front?

The files say no. The fund’s history and money trail are documented in the open: a private donor, Yale as trustee, and 29 years of published grants. It appears nowhere in the MKUltra record, while the real conduits, the Human Ecology Fund and the Geschickter Fund, are named there repeatedly.

How many people died in MKUltra?

No one knows, and anyone who gives you a precise number is guessing. Frank Olson is the one well-documented death connected to the program, and even his case is disputed. Helms had the records destroyed in 1973, which makes a real total impossible to reconstruct. Inflated body counts are usually a sign someone has left the documents behind.

Did anyone go to prison for MKUltra?

No. The records were destroyed before most of the program came to light, the officials involved testified under immunity or not at all, and the courts turned the victims away. James Stanley lost at the Supreme Court in 1987, and the Olson family’s last suit was dismissed in 2013.

Are today’s psychedelic therapies regulated differently?

Very differently. A modern psilocybin or MDMA study runs under an FDA investigational-new-drug application, DEA Schedule I controls, and an institutional review board, with documented informed consent from every participant. Those requirements are the direct legacy of the 1970s reforms. The MDMA case in this article is a reminder that the rules still depend on the people who are supposed to follow them.

Psychiatric Care Grounded in the Evidence

This article is for educational purposes and isn’t medical advice or a substitute for care from your own clinician. The psychedelic treatments it discusses, including psilocybin and MDMA, are investigational, remain federally controlled, and aren’t offered by this practice. If you’re in crisis or thinking about harming yourself, call or text 988 for the Suicide and Crisis Lifeline.

References

  1. Passie T, Seifert J, Schneider U, Emrich HM. The pharmacology of psilocybin. Addict Biol. 2002;7(4):357-364. PMID 14578010. pubmed.ncbi.nlm.nih.gov/14578010
  2. U.S. Senate, Select Committee on Intelligence and Subcommittee on Health and Scientific Research. Project MKULTRA, the CIA’s Program of Research in Behavioral Modification. Joint Hearing, August 3, 1977. intelligence.senate.gov
  3. Commission on CIA Activities Within the United States (Rockefeller Commission). Report to the President. June 1975.
  4. National Security Archive, George Washington University. The John Marks Collection, MKULTRA documents obtained under FOIA. nsarchive.gwu.edu
  5. U.S. House Committee on Oversight and Government Reform, Task Force on the Declassification of Federal Secrets. Mind Control and Accountability: Uncovering the Truth of the CIA’s MKULTRA Experiments. Hearing, June 30, 2026. oversight.house.gov
  6. National Security Archive. MKUltra: Declassification Task Force Should Focus on Real Secrets, Not Conspiracy Theories. Briefing Book #924, June 29, 2026. nsarchive.gwu.edu
  7. Pines M. The Foundations’ Fund for Research in Psychiatry and the growth of research in psychiatry. Am J Psychiatry. 1983;140(1):1-10. PMID 6336636. pubmed.ncbi.nlm.nih.gov/6336636
  8. Kubie LS. Research in Psychiatry Is Starving to Death. Science. 1952;116(3010):239-243. PMID 17742547. pubmed.ncbi.nlm.nih.gov/17742547
  9. Kinzer S. Poisoner in Chief: Sidney Gottlieb and the CIA Search for Mind Control. Henry Holt, 2019.
  10. United States v. Stanley, 483 U.S. 669 (1987). law.cornell.edu
  11. Olson v. United States, 953 F. Supp. 2d 223 (D.D.C. 2013).
  12. The Belmont Report. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 1978 (published in the Federal Register, 1979). U.S. Department of Health and Human Services, Office for Human Research Protections. hhs.gov/ohrp
  13. U.S. Drug Enforcement Administration. Drug Scheduling, Controlled Substances Act. dea.gov
  14. ClinicalTrials.gov. A Randomized, Double-Blind, Controlled Phase 2 Pilot Study of Manualized MDMA-Assisted Psychotherapy. NCT01958593. clinicaltrials.gov
  15. Retraction notes for three MDMA-assisted psychotherapy papers (Mithoefer 2019, Jerome 2020, Feduccia 2021). Psychopharmacology (Berl). 2024;241(11). PMIDs 39126500, 39126501, 39126502. pubmed.ncbi.nlm.nih.gov/39126501
  16. MAPS. Statement: Public Announcement of Ethical Violation by Former MAPS-Sponsored Investigators. May 2019, updated 2022. FDA advisory committee vote and Complete Response Letter reported by Science (June 5, 2024) and NPR (August 9, 2024). Canadian legal and licensing developments reported by CBC News (March 29, 2022; October 16, 2023). Sponsor renaming reported by Psychedelic Alpha (August 28, 2025).
Begin care

Schedule a free consultation

Shantanu Baghel

Board-certified · Psychiatry & forensic psychiatry

Dr. Baghel runs Baghel Psychiatry, a cash-pay telehealth practice serving adults across North Carolina, with a particular focus on the overlap between mood, hormones, and metabolism, and on supervised antidepressant discontinuation.

More writing

Begin

Get a comprehensive evaluation in North Carolina.