Stanford University researchers are pioneering the use of large language models (LLMs) to analyze extensive text data from various sources, including corporate earnings calls and historical documents. This innovative method enables the processing of over 780,000 documents from more than 21,000 companies over a decade, revealing significant insights into economic trends, particularly in response to the tariffs announced by former President Donald Trump in early 2025. The study found that over 30% of earnings calls mentioned negative impacts due to these tariffs, with 60% of calls from U.S. firms reporting adverse effects. Traditional research methods, which often involve labor-intensive surveys and manual data collection, are being outpaced by AI-driven analysis, allowing researchers to generate findings quickly and at scale. This advancement not only enhances the speed and depth of economic analysis but also opens new avenues for interdisciplinary research, demonstrating the transformative potential of AI in understanding complex social and economic dynamics.