When the stakes are higher.
Most of the site treats cannabis as a personal risk decision for an adult. Three situations change that maths: where the risk lands on someone else, on a still-developing brain, or on your livelihood. Here's the straight version of each, so it's an informed choice rather than a blind one.
This isn't scaremongering or telling grown adults what to do. It's that these three areas carry consequences bigger and less reversible than a rough evening, and the honest facts are worth having up front rather than after the fact. For the whole risk picture in one place, see Will it ruin your life?
Pregnancy & breastfeeding
This is the one where "your body, your choice" genuinely stops applying cleanly, because a second person is involved. THC crosses the placenta, so a developing baby is exposed too, and it also passes into breast milk, where, being fat-soluble, it can linger well after use. MIXEDEvidence strength Cannabis use in pregnancy is associated with lower birth weight and possible effects on the baby's brain development; THC also transfers into breast milk and can accumulate. The research is complicated by other factors (tobacco use, other exposures), so exact effect sizes are debated, but the direction of concern is consistent, which is why official guidance is precautionary. Sources: the American Academy of Pediatrics (2018) and the US FDA both advise against cannabis (including CBD) in pregnancy and breastfeeding. UK guidance lands in the same place: NHS advice is to avoid cannabis when pregnant or breastfeeding, and, worth knowing, the Food Standards Agency specifically advises that pregnant and breastfeeding women should not take even shop-bought CBD. If you're pregnant and using cannabis and finding it hard to stop, that's exactly the kind of thing a midwife or GP will help with, without judgement. The Quitting toolkit is here too.
The developing brain
Age genuinely changes the risk. The brain keeps maturing into the mid-20s, and the evidence that cannabis hits a developing brain harder than an adult one is some of the strongest in this whole area. Starting young, and using heavily, is where the real risk sits. STRONGEvidence strength People who start using cannabis in adolescence are markedly more likely to become dependent, roughly 1 in 6 teenage-onset users, versus about 1 in 11 overall, and heavy adolescent use is linked to greater effects on learning, memory and attention. Source: Volkow et al. (2014), NEJM. The much-discussed link between heavy teenage use and lower IQ is more contested (some of it may be explained by other factors) and that debate is covered in Long-Term Effects. The practical upshot isn't complicated: the younger the brain, the higher the stakes, and delaying use (or not using) through the teens and early twenties is one of the genuinely evidence-backed ways to lower the long-term risk. If you're a parent or worried about a young person, Worried about someone? is written for exactly that.
Work & drug testing
Cannabis can follow you into Monday morning in a way alcohol doesn't. THC and its by-products are fat-soluble, so they clear slowly: an occasional user might test positive for a day or two, but a frequent user can show up on a urine test for days to weeks after last using, long after any effect has gone. MIXEDEvidence strength Because cannabis metabolites are stored in fat and released slowly, detection windows vary enormously with how often and how heavily someone uses, from a couple of days for infrequent use to several weeks for daily use. A positive test means "used recently-ish," not "impaired right now." This is the same pharmacology that makes blood-THC a poor marker of recent use in the drug-driving context (Karschner et al., 2009). That matters if your job tests, especially safety-critical roles (driving, machinery, healthcare, aviation) where a positive can cost you the job. A couple of honest points: a prescription doesn't automatically override a workplace drug policy (see Medical cannabis), and the "detox" drinks and kits sold to beat a test are largely dilution tricks: unreliable, and unable to actually clear the fat-stored metabolites. If testing is part of your work, the only dependable approach is time, and the Quitting spoke if stopping is the goal.