Google Research has introduced the Planetary Prediction Engine (PPE), an autonomous AI system that automates the end-to-end geospatial modeling workflow and dramatically accelerates the creation of predictive models for global challenges. The system, part of Google's Earth AI initiative, can execute complex geospatial analytics tasks in minutes rather than the weeks typically required with manual data curation and feature engineering. Given a natural-language query, PPE autonomously retrieves relevant geospatial data, performs feature engineering, trains predictive models, and generates comprehensive reports without human intervention. The system operates through three modular stages orchestrated by large language models: intelligent geospatial data selection that retrieves relevant covariates from established repositories and live web sources, multimodal dataset curation that fuses data with pre-trained foundation model embeddings while implementing automated safeguards against target leakage, and automated model building that optimizes across multiple model families while preventing overfitting. Google demonstrated PPE's effectiveness across diverse domains, achieving significant improvements over manual expert pipelines. On CDC health indicators, the system achieved a mean R² of 76.8% compared to 60.0% for manual approaches; for food security modeling in Nigeria, PPE doubled baseline accuracy when downscaling vulnerability assessments from regional to local government levels. The breakthrough addresses a critical bottleneck in planetary-scale analytics, particularly important during humanitarian crises where rapid modeling is essential. By automating specialized geospatial workflows that previously required interdisciplinary teams, PPE accelerates decision-making in public health, food security, environmental risk assessment, and socioeconomic mapping, demonstrating how AI orchestration can translate complex domain workflows into scalable, reproducible systems.