Amazingly, almost 50 years after the first demonstration of anticancer effects of cannabinoids in vitro and in vivo, well-designed clinical trials that definitively prove tumour-inhibiting effects in man are still missing. Whereas a large number of preclinical studies exist that describe tumour-inhibiting effects of cannabinoids, alone or in combination, but also in the form of medical cannabis or natural extracts in vitro, the number of in vivo studies is still limited. Even more limited are well-documented experiences in man. Most animal studies and experience with cannabinoids in man concern brain tumours. This review summarises the effects of phytocannabinoids in brain, breast, colorectal, head and neck, haematological, liver, lung, pancreatic, ovarian, prostate, and skin cancers in animal models and, if available, in patients. The large majority of animal studies demonstrate tumour-inhibiting effects of cannabinoids, thus confirming in vitro data. Experiences in cancer patients are almost exclusively limited to individual case reports and case series without a control group. Many questions are currently unanswered such as the role of pure cannabinoids compared to combinations, cannabinoids as the eventual sole cancer therapy, optimal dosages, or duration of treatment. Pure cannabidiol (CBD) seems to be superior to pure delta-9-tetrahydrocannabinol (THC) in experimental settings. The role of medical cannabis or extracts is less clear as they vary in their phytochemical composition. In conclusion, cannabis/cannabinoids may slow the progression of tumours. However, the hope that cannabinoids could eventually cure cancer as often spread in social media, is, at present, wishful thinking. Above all, well-designed clinical trials paired with long-term follow-up of cancer patients are needed.