Counterintuitively, training AI models with data from disparate physical domains, like mining, improves the performance of systems in completely different areas, such as self-driving cars. This cross-domain learning suggests that a broad understanding of the physical world is key to robust, real-world AI.

Unlike traditional neural networks which require fixed-resolution inputs (e.g., pixels), neural operators model data as continuous functions. This allows them to "zoom in" and make predictions at resolutions higher than the training data, a crucial capability for multi-scale physical phenomena like weather patterns.

Counter-intuitively, successful weather and climate AI models are not trained on long-term data. They are trained to predict only the next six hours autoregressively. This surprisingly generalizes to stable rollouts predicting weather patterns for hundreds or thousands of steps into the future, enabling long-term forecasting from short-term training.

In a battle of methods, Natera's deep learning AI, trained on millions of samples classified by classical statistical models, began to outperform its teachers. The AI was better at identifying the underlying noise and difficult outlier cases, demonstrating a non-obvious capability of AI to find patterns beyond its explicit training logic.

A key risk in deploying AI is its inability to generalize to 'long-tail' or out-of-distribution events. Models trained on vast but finite data often fail when encountering novel situations common in the open-ended real world, such as a self-driving car mistaking a stop sign on a billboard for a real one.

Current AI can learn to predict complex patterns, like planetary orbits, from data. However, it struggles to abstract the underlying causal laws, such as Newtonian physics (F=MA). This leap to a higher level of abstraction remains a fundamental challenge beyond simple pattern recognition.

Traditional weather forecasting requires massive supercomputers, limiting access to large agencies. New AI models are tens of thousands of times faster and can run on a single consumer-grade GPU. This democratizes high-fidelity weather modeling for smaller agencies and nations, especially in the global south.

Most AI weather models project the Earth onto a flat rectangle, causing simulations to become unstable and "blow up" over long periods. By incorporating the planet's spherical geometry using Fourier Neural Operators, models like ForecastNet remain stable for long-term climate rollouts, effectively becoming climate models.

By training on data across many cancer types ("pan-cancer"), AI models learn universal biological principles. This approach allows them to generalize learnings from large, common cancer datasets to significantly improve prediction accuracy for rare cancers, which often suffer from a lack of specific data for training effective models.