What the AKT1 and COMT gene research actually shows about cannabis-induced psychosis risk
Why teen cannabis use carries different risk than adult use, beyond “developing brains are more sensitive”
How today’s THC potency compares to cannabis from the 1980s and 90s
How cannabis-induced psychosis risk compares to psychedelic-induced psychosis risk, and why the gap matters
Who Johnny’s Ambassadors and POCCIP are, and why parent advocacy groups are opposing federal rescheduling
What to actually watch for if you’re a parent, or reflecting on your own teen cannabis use
A few Notes I published this spring on cannabis and mental health pulled in more comment traffic than almost anything else I’ve written this year, including personal testimony, pushback on the research I cited, and something I want to name directly: parents whose teenagers developed psychosis after using high-potency THC, now organizing against federal rescheduling because of what happened to their kids.
That group deserves a real answer, not a dismissal. So does everyone else asking a version of the same question: is there a gene that makes some teenagers vulnerable to cannabis-induced psychosis? Here is where the research stands, including how it compares to the psychedelics conversation happening in the same comment threads.
Yes, and it has a name. Two genes show up consistently in the research, AKT1 and COMT, both involved in dopamine signaling, the same system implicated in schizophrenia and psychosis.
The clearest work comes from Dr. Marta Di Forti’s team at King’s College London. In a case control study of 489 first-episode psychosis patients and 278 healthy controls, carrying a specific AKT1 variant (rs2494732) roughly doubled the probability of psychosis in cannabis users, and daily use pushed that risk up to sevenfold. A separate UCL study of 442 young cannabis users, tested intoxicated and again sober, found that people with this variant reported more psychotic-like symptoms (paranoia, visual distortion) while high, even without a family history of schizophrenia. COMT shows a similar pattern, though its findings have been harder to replicate.
What this doesn’t mean: a genetic test you can buy that tells you whether your kid is safe to use cannabis. That test doesn’t exist. What it does mean is that “cannabis causes psychosis” and “cannabis has zero role in psychosis” are both wrong. The honest version is narrower: cannabis is a risk factor whose size depends heavily on who is using it, and genetics is one variable determining who.
Three things compound rather than add.
First, timing. The adolescent brain is still building the neural architecture involved in impulse control, threat appraisal, and emotional regulation, and THC interacts directly with the endocannabinoid system doing that building. Research on chronic cannabis exposure in early life points to changes in serotonin 2A receptor signaling as one mechanistic pathway connecting adolescent use to later psychosis-like symptoms, distinct from the pathway seen in adults.
Second, potency. This is where the “it’s just weed” framing stops applying. Flower that measured 1 to 2 percent THC in the 1980s now regularly measures 20 to 30 percent, and concentrates can run close to 90 to 99 percent THC. The dose-response relationship is not subtle: heavier, more frequent use of higher-potency product raises risk in a graded way, not an on-off way.
Third, age of onset. Schizophrenia typically presents in the late teens to early twenties for men and late twenties to early thirties for women, and cannabis use is associated with earlier onset, by as much as several years in some analyses. Earlier onset generally means a harder course of illness.
None of this means most teenagers who use cannabis will develop psychosis. They won’t. But “most people are fine” and “the risk is not evenly distributed” are both true at once, and public conversation keeps treating them as if only one can be.
Short version: the risk profile is real but smaller and differently shaped than cannabis’s. Cannabis carries the higher transition rate to schizophrenia (roughly 34 percent versus 13 to 26 percent for hallucinogens), the identified genetic pathway, and a chronic dose-response mechanism that psychedelics don’t share. Full comparison, including the incidence data and why the mechanisms differ, is in the FAQ below.
This is the part of the comment threads I want to take seriously rather than wave off.
Laura Stack founded Johnny’s Ambassadors after her 19-year-old son Johnny died by suicide in 2019, following a psychotic break she attributes to dabbing high-potency THC concentrate he began using at 14. Her organization now runs Parents of Children with Cannabis-Induced Psychosis, a support group with several thousand members. When the administration moved to reschedule cannabis from Schedule I to Schedule III this year, Stack and groups like hers opposed it publicly, arguing rescheduling will read to teenagers as a safety signal the evidence doesn’t support, regardless of what it actually changes legally.
I understand the instinct in cannabis and psychedelic spaces to read that opposition as reflexively anti-legalization. I’d push back. These are parents responding to documented harm to their own children, and the products driving that harm (concentrates, high-potency vapes, dabs) are largely downstream of an unregulated market, not the plant at the potencies it existed a generation ago. Their argument isn’t “cannabis is evil.” It’s “the products on shelves today aren’t the products the safety conversation was built around, and our kids paid for that gap.” That’s a harm reduction argument, even when it doesn’t sound like one from a media space that mostly talks about access.
Where I disagree with some of these groups is the claim that there’s “zero evidence” any THC product helps anyone. Overstating the case in that direction doesn’t serve teenagers either. The evidence for dose-dependent harm to developing brains is strong, and the evidence for benefit in other contexts and populations is also real. Both can be true because they describe different products, different ages, different nervous systems.
Watch for actual warning signs rather than treating all teen cannabis use as equally dangerous. Frequency, potency, and age of first use are the variables the research keeps pointing back to, not use itself. A teenager dabbing concentrate daily is a different risk profile than one who tried an edible twice at a party.
Family history matters. A first-degree relative with schizophrenia, bipolar disorder with psychotic features, or a documented psychotic episode changes the risk calculus for a teenager, in a way that deserves a direct conversation, not a vague one.
The takeaway I keep landing on: the fight over rescheduling and the fight over teen mental health are being treated as the same argument, and they aren’t. You can support adult access to a regulated, well-labeled cannabis market and still think a 15-year-old shouldn’t be dabbing 90 percent THC concentrate. Pretending those positions are in tension is what’s failing both the parents in my comments section and the researchers trying to get anyone to fund the studies that would settle this.
If a piece of this connects to something you’re navigating, feel free to use me as a resource.
Take care,
April
Does cannabis cause psychosis, or just increase the risk? Cannabis is a risk factor for psychosis, not a direct cause. Most people who use cannabis, including most teenagers, never develop a psychotic disorder. Risk rises with genetic vulnerability, potency, frequency of use, and age of first use, and the size of that risk differs sharply from person to person.
What gene is linked to cannabis-induced psychosis? AKT1 is the gene with the most consistent research behind it. A specific AKT1 variant (rs2494732) has been associated with roughly double the risk of psychosis in cannabis users, rising to about sevenfold with daily use. COMT has also been studied for a similar link, though those findings have been harder to replicate across research groups.
Is there a genetic test for cannabis psychosis risk? No. There is no clinically available test that tells an individual, including a teenager, whether they carry elevated genetic risk for cannabis-induced psychosis. The AKT1 and COMT research comes from population studies, not individual diagnostic screening.
Is cannabis or psychedelics riskier for psychosis? Cannabis carries a higher risk by the clearest available measure: roughly 34 percent of people with cannabis-induced psychosis go on to develop schizophrenia, compared to an estimated 13 to 26 percent for hallucinogens like psilocybin and LSD. In controlled clinical trial settings, psychedelic-induced psychosis is rare (0.002 to 0.6 percent depending on study type), rising sharply outside supervision, as shown by a 2024 Ontario study linking hallucinogen-related ER visits to a 3.5-fold increased risk of later schizophrenia after adjusting for other factors. The mechanisms also differ: cannabis risk builds through chronic, dose-dependent, unsupervised use interacting with an identified genetic pathway (AKT1), while psychedelic-induced psychosis is typically tied to a single acute exposure in an already-vulnerable person, and no equivalent risk gene has been identified for psychedelics.
Why are parents opposing cannabis rescheduling? Groups like Johnny’s Ambassadors, founded by Laura Stack after her son’s death by suicide following cannabis-induced psychosis, argue that rescheduling cannabis from Schedule I to Schedule III sends teenagers a safety signal the evidence doesn’t support, particularly around high-potency concentrates and vapes. Their opposition is rooted in documented harm to their own children, not a rejection of adult access broadly.
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.