The Psychedelic Hype vs. the Evidence
“The Psychedelic Revolution Is Coming.” “How Magic Mushrooms Can Fix Depression.” “Net-zero trauma by 2070.” The psychedelic hype is a decade old and backed by billions in investor money. The evidence underneath it is a handful of small trials that nobody could blind, one drug the FDA has already rejected, and a set of harms the headlines leave out. Here is the claim, then the data, with every figure linked to its source.
Every pooled analysis of psilocybin for depression finds an effect. Every one also finds bias, broken blinds, industry conflicts and shrinking effects in bigger trials. No classic psychedelic is FDA-approved. MDMA was rejected.
What is being sold, and what the data show
Five years on, the FDA has approved no classic psychedelic for any condition. The one that reached a decision, MDMA for PTSD, was voted down 9 to 2 and 10 to 1 by the FDA’s advisers and rejected in August 2024. NYT, via MAPS; FDA hearing summary, September 2026
The largest psilocybin trial ever run, 581 patients, found a 3.8-point difference on a 60-point depression scale at six weeks. The first Phase 3 found 3.6 points. The company’s own Phase 2b found no significant sustained response at 12 weeks. That is a modest antidepressant signal, not a cure. Psychology Today; Huck; Compass, July 2026; Compass, SEC filing, June 2025; Goodwin et al., NEJM, 2022
A year later the FDA rejected MDMA-assisted therapy, three of the group’s papers were retracted for “protocol violations amounting to unethical conduct,” and the company cut 75 percent of its staff and parted ways with Doblin. The Microdose, 2023; STAT, August 2024; GeneOnline, August 2024
True at three weeks: 29 percent remission on 25 mg versus 8 percent on the 1 mg control. By week 12 the sustained-response difference was no longer statistically significant, and suicidal ideation, behavior or self-injury occurred in every dose group. The authors themselves wrote that suicidality “demands clinical vigilance.” The Guardian, November 2022; NEJM, 2022; HCPLive
As of October 2026 neither Compass Phase 3 trial has been published in a peer-reviewed journal; both exist as press releases and webcasts. The company’s stated launch target is 2027, “subject to FDA approval and DEA rescheduling,” with $433 million in cash waiting on it. Compass, July 2026; Compass Q2 2026 release, SEC
Testimonials are the oldest form of drug marketing. In the largest survey of bad psilocybin trips, 84 percent said they benefited, and in the same survey 11 percent put themselves or others at risk of physical harm, 7.6 percent sought treatment for enduring symptoms, and three people attempted suicide. Both things are true, which is why medicine runs trials instead of collecting stories. ITV, March 2023; Carbonaro et al., Johns Hopkins, 2016
A market projection is a forecast of sales, not of cures. The companies making it hold the patents: Compass holds five U.S. patents on synthetic psilocybin, raised more than $300 million in its IPO, and an investor email bragged that rivals “will never be able to bring a product to market as they will hit the patents of Compass and ATAI.” High Times, 2023; The American Prospect, 2022
The effect shrinks as the trials get bigger and better
Psilocybin vs. control on the MADRS depression scale (60 points)
The headline studies were tiny. The 24-person Johns Hopkins trial had no placebo at all, just a waiting list. The two Phase 3 trials, with ten to twenty times the patients, found about half the Phase 2b effect. A 2025 meta-analysis put it plainly: “Smaller studies evidenced stronger effects” and effects “attenuated in larger and better-controlled studies.”
Davis et al., JAMA Psychiatry, 2021; Raison et al., JAMA, 2023; Goodwin et al., NEJM, 2022; Compass, June 2025; Compass, July 2026; Psychopharmacology, 2025
Head to head with an ordinary antidepressant
“This trial did not show a significant difference in antidepressant effects between psilocybin and escitalopram.”
Carhart-Harris et al., New England Journal of Medicine, April 2021
The one trial that compared psilocybin directly with a generic SSRI found no significant difference on the primary endpoint (P = 0.17). The secondary outcomes that favored psilocybin, the authors noted, “lacked correction for multiple comparisons.”
What the pooled analyses say
BMJ 2024, 436 participants: a moderate effect, “all studies had a moderate risk of bias,” and the journal attached an expression of concern. Psychopharmacology 2025, nine trials: “Most studies demonstrated a high risk of bias” and “almost all studies documented financial conflicts of interests.” JAMA Network Open 2025: control groups in psilocybin trials improved far less than in SSRI or esketamine trials, which “might indicate that psilocybin’s antidepressant efficacy is overestimated.” Cochrane 2024: “blinding could not be achieved,” certainty “low to very low.”
Metaxa and Clarke, BMJ, 2024; Psychopharmacology, 2025; JAMA Network Open, 2025; Cochrane, 2024
The honest summary: psilocybin probably does something for depression in the short term. So does escitalopram, which costs pennies and does not require two therapists and a six-hour supervised session. Whether the something is worth the risks is exactly what the FDA exists to decide, and it has not decided yet.
Everyone knows who got the drug. That is not a detail.
MDMA trials: who guessed their assignment correctly
The FDA’s own briefing: “It is reasonable to assume that functional unblinding and expectation bias has impacted treatment effects observed in the clinical trials with MDMA to some extent.” When Stanford researchers gave ketamine to patients under general anesthesia, so no one could feel the trip, the drug did no better than placebo.
FDA briefing document, June 2024; Lii et al., Nature Mental Health, 2023
“Placebo is probably the single most effective, consistent intervention in medicine, full stop.”
Boris Heifets, MD, PhD, Stanford Medicine, October 2023
“Treatment effect sizes in psychedelic RCTs are likely over-estimated due to de-blinding of participants and high levels of response expectancy.”
Muthukumaraswamy et al., Expert Review of Clinical Pharmacology, 2021
In July 2026 the FDA made it official. Its final guidance on psychedelic trials says functional unblinding makes a traditional placebo “problematic for assessing efficacy,” that results must be “strongly persuasive and robust across study endpoints to overcome biases,” and that prior psychedelic users are especially likely to recognize the drug and “anticipate a treatment benefit.”
FDA, Psychedelic Drugs: Considerations for Clinical Investigations, final guidance, July 2026
The hype’s flagship went to the FDA. Here is what happened.
- 2021 and 2023. Two MAPS-funded Phase 3 trials, 90 and 104 patients, report large effects (67 and 71 percent of MDMA patients no longer meeting PTSD criteria). Blinding “was not formally assessed.” Mitchell et al., Nature Medicine, 2021; 2023
- June 4, 2024. FDA advisers vote 9 to 2 that effectiveness was not shown and 10 to 1 that benefits do not outweigh risks. The chair: “I felt the functional unblinding, the lack of management for expectation bias.” Managed Healthcare Executive
- June 27, 2024. ICER rates the evidence “insufficient,” citing “functional unblinding in the trials and additional concerns around trial design and conduct,” and declines to set a price. ICER
- August 9, 2024. FDA rejects the application and asks for another Phase 3 trial. Drug Topics
- August 10 to 16, 2024. Psychopharmacology retracts three MDMA papers for “protocol violations amounting to unethical conduct” at a trial site. Lykos cuts 75 percent of its staff and parts with Rick Doblin. STAT; GeneOnline
The full story, including the therapist abuse case that triggered the retractions, is on our MDMA therapy and the FDA page.
The risks that do not make the headline
Poison center calls, psilocybin, ages 13 to 25
4,055 exposures from 2013 to 2022. “Cases started increasing in 2019. In 2022, cases more than tripled among adolescents and more than doubled among young adults, compared to 2018.” Three quarters of single-substance adolescent cases needed medical attention. The authors note the study was not designed to measure decriminalization’s effect; 2019 is the year Denver decriminalized.
Farah et al., Journal of Adolescent Health, 2024
Psychosis
Among 9.2 million Ontarians, hallucinogen-related emergency visits rose 86 percent from 2013 to 2021. People with such a visit had 3.5 times the risk of a schizophrenia-spectrum diagnosis within three years after full adjustment (21 times before adjusting for other substance use and mental health history): 4.0 percent versus 0.15 percent.
Lasting effects
Among 608 people who reported extended difficulties after a psychedelic, “for approximately one-third of the participants, problems persisted for over a year, and for a sixth, they endured for more than three years” (a self-selected sample with no denominator). Hallucinogen persisting perception disorder is “an uncommon yet serious event”; in a 2025 naturalistic cohort more than 30 percent reported HPPD-type effects at four weeks, rarely distressing. In the Compass Phase 2b trial, suicidal ideation, behavior or self-injury occurred in every dose arm.
Evans et al., PLOS ONE, 2023; Expert Opinion on Drug Safety, 2022; PNAS Nexus, 2025; Goodwin et al., NEJM, 2022
Outside the clinic
Matthew Perry died in October 2023 of “the acute effects of ketamine” that the medical examiner said could not have come from his clinic infusion a week and a half earlier. The “clinic” framing does not stay in the clinic. In Oregon’s legal program, 1.2 percent of 346 clients, all first-time users, had adverse reactions requiring medical attention; the same study reported mental health improvements, which we note.
OPB, December 2023; Korthuis et al., JAMA Network Open, 2026, via Psych Congress Network
Drug by drug: Is psilocybin safe? · MDMA therapy and the FDA · Ibogaine treatment risks
Who profits if the hype wins
Compass Pathways holds five U.S. patents on synthetic psilocybin, raised more than $300 million in its 2020 IPO with Peter Thiel among its backers, and reported $433 million in cash in August 2026 with a launch targeted for 2027. Lykos (formerly MAPS PBC) raised “more than $100 million in an oversubscribed Series A” seven months before the FDA rejected its drug. The trade press projects an $11.8 billion market by 2029.
The American Prospect, 2022; Compass, SEC, August 2026; MAPS, January 2024; High Times, 2023
The science is funded by the sellers. A 2025 meta-analysis: “Almost all studies documented financial conflicts of interests. Independent RCTs from researchers without financial conflicts of interest are needed.” A 2024 review warned that corporate funding creates “a potential rise in conflicts of interest in relevant studies and publications.”
And now the politics. An April 2026 executive order directs priority vouchers, a Right to Try pathway for “psychedelic drugs, including ibogaine compounds,” at least $50 million for state partnerships and a rescheduling review. FDPS’s response: “This order elevates politics and hype over sound science.”
Psychopharmacology, 2025; Koning, Solmi and Brietzke, 2024; Executive Order, April 18, 2026; FDPS statement, April 2026; STAT, April 2026
The field’s own founders are warning about the hype
“I am concerned that there are some totally unrealistic expectations of benefits. There are real risks that can result in injury, death, or onset of new debilitating psychiatric conditions. I count myself among the very conservative with respect to the movements afoot to legalize.”
Roland Griffiths, PhD, founder of the Johns Hopkins Center for Psychedelic and Consciousness Research, Johns Hopkins Magazine, 2023
“If people can guess the treatment allocation, then this is functional unblinding, and it is a real problem.”
Charles Raison, MD, lead investigator of the Usona psilocybin trial, Psychedelic Alpha, 2025
“Psychedelic science is facing serious challenges that threaten the validity of core findings and raise doubt regarding clinical efficacy and safety.”
Michiel van Elk and Eiko Fried, Leiden University, Therapeutic Advances in Psychopharmacology, 2023
Matthew Johnson, who left the Hopkins center, has said it ran like a “new-age retreat center” with a “cult-like atmosphere,” and that the trial results cannot validate the mystical claims made for them. The regulator agrees with the skeptics: in September 2026 the FDA held a public hearing on psychedelics at which speakers from Oxford and Dartmouth testified that real-world risk is “more likely to be underestimated than overestimated” and that “certainty of evidence remains low.” Who else opposes psychedelic legalization.
Reason, February 2025; FDA public hearing, September 14, 2026
Psychedelic hype: common questions
Are psychedelics FDA-approved?
No classic psychedelic is FDA-approved for any condition as of October 2026. MDMA-assisted therapy for PTSD was rejected in August 2024 after advisers voted 9 to 2 that effectiveness was not shown. Compass Pathways is submitting a psilocybin application on a rolling basis, with a decision not expected before 2027. Esketamine (Spravato), approved in 2019, is a ketamine derivative dispensed under a restricted program, not a classic psychedelic.
Does psilocybin cure depression?
No. The best trials show a modest short-term antidepressant effect: 3.6 to 3.8 points on a 60-point scale at six weeks in the two Phase 3 trials, about half the earlier Phase 2b result, and no significant difference from the generic SSRI escitalopram in the one head-to-head trial. Pooled analyses find a real but moderate effect, high risk of bias and conflicts of interest in almost every study.
Why does blinding matter in psychedelic trials?
Because a trial only proves a drug works if patients and raters do not know who got it. In the MDMA trials roughly 90 percent of drug recipients guessed correctly, and the FDA’s own briefing said expectation bias had affected the results. When ketamine was given under anesthesia so no one felt it, it did no better than placebo. The FDA’s 2026 guidance now requires results ‘strongly persuasive and robust’ enough to overcome this bias.
What happened with MDMA therapy and the FDA?
FDA advisers voted 9 to 2 in June 2024 that MDMA-assisted therapy was not shown to be effective for PTSD and 10 to 1 that benefits did not outweigh risks, citing functional unblinding and trial conduct. ICER rated the evidence insufficient. The FDA rejected the application in August 2024 and asked for another Phase 3 trial; three of the sponsor’s papers were retracted days later for unethical conduct at a trial site.
What are the risks of psychedelics?
Bad trips (11 percent of people in a large survey put themselves or others at physical risk), persisting perceptual disorder, lasting psychological difficulties that in a self-selected sample persisted more than a year for a third of respondents, suicidality signals in the Compass depression trial, and psychosis: people with a hallucinogen-related emergency visit in Ontario had 3.5 times the adjusted risk of a schizophrenia-spectrum diagnosis within three years. Psilocybin poison center calls among teens more than tripled from 2018 to 2022.
Does FDPS oppose psychedelic research?
No. FDPS supports rigorous, independent, replicated trials and FDA approval of anything that passes them, which is how every real medicine reaches patients. What FDPS opposes is medicine by ballot measure, executive order or marketing campaign, and the commercialization of Schedule I drugs before the science is done.
Related: Arguments against psychedelic legalization · Who opposes psychedelic legalization · Is psilocybin safe? · MDMA therapy and the FDA
Trials, not hype.
If psychedelics are medicine, they should be proven like medicine. FDPS has the evidence, the policy positions and the model language for lawmakers who are being told otherwise.
