A study published in *Scientific Reports* introduces a Hybrid AI and OCPP framework for city-scale electric vehicle charging, linking forecasting, scheduling, and anomaly detection into one pipeline. The framework uses a short-horizon forecasting engine combining Prophet, XGBoost, and GRU, a tariff-aware scheduling policy with a fairness guardrail, and a lightweight anomaly module. Decisions are encoded as OCPP 1.6 and 2.0.1 operations. On a two-year, multi-station dataset with 1,553 operational days and 455 days with price coverage, the scheduler reduced feeder peak and charging cost by 5.0% with GA and by 8.2% with GA + Q on average. For anomaly detection, a CNN reached ROC AUC 0.914, while an Autoencoder and Isolation Forest provided lower AUC. All modules run in a rolling 15-minute loop with optimization overhead below 0.1 s. The study addresses the need for standards-based deployment at city scale with measured latency and anomaly handling, focusing on reproducible peak and cost reductions under posted tariffs and daily peak caps.