The research team from UC Berkeley has made significant breakthroughs in the field of prediction with their developed language model system, which has approached or even surpassed human average levels. By utilizing large-scale data and rapid processing capabilities, they have automated key prediction processes, demonstrating high potential for accuracy. The results show that the system's performance on the test set is close to human levels, providing important references for improving prediction precision and efficiency. This research offers strong support for integrating language models into the prediction field, showcasing their potential in enhancing prediction accuracy and efficiency.