Researchers at the University of Tokyo applied a machine‑learning predictor for insulin resistance (AI‑IR) to 500,000 UK Biobank participants and found insulin resistance is a statistically significant risk factor for 12 cancer types. The results were published in Nature Communications and reported by the study authors led by Yuta Hiraike, MD, PhD. The team used nine routine clinical measurements to estimate insulin resistance across a large cohort and then correlated those estimates with cancer incidence. The lead finding reports a broad metabolic–oncology connection at population scale, implicating insulin resistance beyond its established roles in diabetes and cardiovascular disease. The work may prompt researchers and clinicians to consider insulin‑resistance screening in oncology risk models and to explore metabolic interventions as adjuncts to cancer prevention. Insulin resistance is a condition where tissues respond poorly to insulin, raising blood glucose and altering metabolic signaling pathways; AI‑IR is a predictive model trained to infer that state from standard clinical labs.