A new AI model called COMPASS, developed by researchers at Harvard Medical School led by Marinka Zitnik, was reported to improve prediction of patient response to immune checkpoint inhibitors (ICIs). The work positions COMPASS as a tool for anticipating which patients may benefit from therapy. The announcement underscores the shift from purely retrospective biomarkers toward computational models that attempt to integrate heterogeneous clinical and biological inputs. If validated, COMPASS-type systems could support treatment selection and stratification decisions across ICI regimens. For biotech professionals, the key point is that the COMPASS framework is explicitly designed for immune checkpoint response forecasting, aligning with how companies are increasingly packaging AI alongside clinical development to de-risk responder enrichment strategies.