The study presented by Fan et.al shows a way for developing highly effective and minimally toxic AR inhibitors, however further study is required to check the long-term safety. The researchers built a machine learning virtual-screen model using androgen receptor activity data from ChEMBL and combined it with molecular docking to identify potential AR inhibitors. The top 20 compounds were experimentally tested through cell-based assays and molecular analyses to evaluate their inhibitory effects and therapeutics potential. The researchers used a random forest model and molecular docking to virtually screen ChemDiv’s library. This found 20 candidate androgen receptor inhibitors, of which 8020-1599 and C301-6562 showed the strongest activity. These two compounds significantly inhibited prostate cancer cell proliferation, migration, and invasion in vitro, with effects comparable to Enzalutamide. In vivo, both compounds suppressed tumor growth without toxicity and blocked androgen receptor signaling by preventing AR nuclear translocation.This study showed that combining computer-based screening with lab testing is an effective way to find new drugs that block the androgen receptor (AR). The two main compounds, 8020-1599 and C301-6562, strongly slowed tumor growth with little to no toxicity by preventing AR activity and its related signaling.