Docking with the International Space Station (ISS) is a complex process, much like trying to parallel park a car while traveling at high speed in space. Researchers at Stanford University have developed a new artificial intelligence model called the Out-of-this-World-Model (OWM) to improve the accuracy of spacecraft docking. Unlike traditional methods that rely on guidance, navigation and control (GNC) algorithms and extended Kalman filters, the OWM uses mental simulations to predict and adapt to docking scenarios in real time. These traditional systems often struggle with the high-speed video processing required for precise docking maneuvers. In space, traditional computer vision techniques can be unreliable due to environmental factors like sunlight reflecting off solar panels. Reinforcement learning (RL) algorithms, which are commonly used in controlled environments, also face challenges when conditions change, such as when the ISS docking port moves. The OWM model, however, learns from experience rather than relying on pre-defined equations. It simulates potential future scenarios to guide spacecraft more effectively. To train the OWM model, researchers used a software library called AstroJAX, which runs on graphics processing units (GPUs). This significantly reduced the training time compared to traditional methods. The OWM model required 500,000 iterations to master docking maneuvers, a fraction of the 25,000,000 iterations needed for a comparable RL system. It performed better in new and unexpected docking situations, successfully docking at about 53% of the ISS ports, compared to 29% for the RL algorithm. Despite its improvements, the OWM model still faces challenges, particularly in close-up operations where the risk of collision is high and heavily penalized. Future versions of the model could address these issues. The success of the OWM highlights the potential for advancements in autonomous spacecraft operations. However, achieving full automation, especially in scenarios where humans are present, may take more time to develop and implement.