Cannabidiol (CBD) is being investigated as a treatment for several medical disorders but there is uncertainty about its safety. We conducted the first systematic review and meta-analysis of the adverse effects of CBD across all medical indications. Double-blind randomized placebo-controlled clinical trials lasting ≥7 days were included. Twelve trials contributed data from 803 participants to the meta-analysis. Compared with placebo, CBD was associated with an increased likelihood of withdrawal for any reason (OR 2.61, 95% CI: 1.38–4.96) or due to adverse events (OR 2.65, 95% CI: 1.04–6.80), any serious adverse event (OR 2.30, 95% CI: 1.18–4.48), serious adverse events related to abnormal liver function tests (OR 11.19, 95% CI: 2.09–60.02) or pneumonia (OR 5.37, 95% CI: 1.17–24.65), any adverse event (OR 1.55, 95% CI: 1.03–2.33), adverse events due to decreased appetite (OR 3.56, 95% CI: 1.94–6.53), diarrhoea (OR 2.61, 95% CI: 1.46–4.67), somnolence (OR 2.23, 95% CI: 1.07–4.64) and sedation (OR 4.21, 95% CI: 1.18–15.01). Associations with abnormal liver function tests, somnolence, sedation and pneumonia were limited to childhood epilepsy studies, where CBD may have interacted with other medications such as clobazam and/or sodium valproate. After excluding studies in childhood epilepsy, the only adverse outcome associated with CBD treatment was diarrhoea (OR 5.03, 95% CI: 1.44–17.61). In summary, the available data from clinical trials suggest that CBD is well tolerated and has relatively few serious adverse effects, however interactions with other medications should be monitored carefully. Additional safety data from clinical trials outside of childhood epilepsy syndromes and from studies of over-the-counter CBD products are needed to assess whether the conclusions drawn from clinical trials can be applied more broadly.