Abstract
Investing in cancer prevention can be cost-effective. However, this requires significant changes both inside and outside the health care system. The core of the preventive strategy is the assignment of an individual risk level of developing cancer. Artificial intelligence (AI), which has emerged as a tool to reduce errors and confusion in data collection and analysis, has helped accelerate recent advances in identifying circulating markers to generate predictive methods. With predictive models applied to increasingly less invasive and repeatable analytic tests, the risk is no longer assigned but profiled directly on the individual over time. On this basis, the probability of early cancer diagnosis is increased and at the same time, proactive preventive medicine transits from offering lifestyle recommendations to guiding specific treatments to reduce the risk. Despite these promises, AI-based predictive models also present challenges in clinical implementation. Addressing these challenges is crucial to minimizing the future burdens associated with fighting cancer.