As countries adopt more permissive cannabis policies, it is increasingly important to identify strategies that can reduce the harmful effects of cannabis use. This study aimed to determine if increasing the CBD content of cannabis can reduce its harmful effects. Forty-six healthy, infrequent cannabis users participated in a double-blind, within-subject, randomised trial of cannabis preparations varying in CBD content. There was an initial baseline visit followed by four drug administration visits, in which participants inhaled vaporised cannabis containing 10 mg THC and either 0 mg (0:1 CBD:THC), 10 mg (1:1), 20 mg (2:1), or 30 mg (3:1) CBD, in a randomised, counter-balanced order. The primary outcome was change in delayed verbal recall on the Hopkins Verbal Learning Task. Secondary outcomes included change in severity of psychotic symptoms (e.g., Positive and Negative Syndrome Scale [PANSS] positive subscale), plus further cognitive, subjective, pleasurable, pharmacological and physiological effects. Serial plasma concentrations of THC and CBD were measured. THC (0:1) was associated with impaired delayed verbal recall (t(45) = 3.399, d = 0.50, p = 0.001) and induced positive psychotic symptoms on the PANSS (t(45) = -4.709, d = 0.69, p = 2.41 × 10<sup>-5</sup>). These effects were not significantly modulated by any dose of CBD. Furthermore, there was no evidence of CBD modulating the effects of THC on other cognitive, psychotic, subjective, pleasurable, and physiological measures. There was a dose-response relationship between CBD dose and plasma CBD concentration, with no effect on plasma THC concentrations. At CBD:THC ratios most common in medicinal and recreational cannabis products, we found no evidence that CBD protects against the acute adverse effects of cannabis. This should be considered in health policy and safety decisions about medicinal and recreational cannabis.