Researchers unveiled an AI system intended to deliver “just-in-time,” risk-stratified evaluations of sepsis in neonatal intensive care units (NICUs). The approach is designed to support earlier detection and management by tailoring assessments to a patient’s evolving risk profile rather than using static or delayed decision points, according to the report. Sepsis remains a major driver of neonatal morbidity and mortality, and early recognition is often limited by clinical variability and the time needed for confirmatory testing. By prioritizing risk scoring inside the care workflow, the model targets a practical bottleneck in NICU operations. If validated across settings and integrated safely into clinical decision-making, the system could change how teams triage suspected infection and allocate monitoring resources. Adoption will likely depend on performance across diverse patient populations, transparency around model drift, and clear protocols for clinician actioning.