Technology executives are making bold claims that artificial intelligence will cure cancer within five to ten years or within lifetimes, yet there is considerable tension between these grandiose proclamations and the measured progress of actual scientific research. Whilst AI has demonstrated genuine applications in cancer detection, prevention and treatment—including clinical use in lung imaging and brain tumour removal—medical experts caution that such sweeping promises underestimate the complexity of the challenge and the limitations of current technology.

Cancer is not a single disease but rather hundreds of distinct conditions determined by location and genetic factors, each with different characteristics and treatment requirements. Medical oncologists acknowledge that AI will likely improve diagnostic efficiency and enable more personalised care, but emphasise that without AI systems capable of thinking and conducting experiments autonomously at scale, a universal cure across all cancer types remains implausible in the near term. The complexity is exemplified by the brain alone, which contains over 100 distinct tumour types, highlighting the fundamental scientific obstacles to any singular solution.