Imagine an AI that could predict your next job based on your career history. That’s the promise of LABOR-LLM, a new approach using large language models (LLMs) to forecast career trajectories. Traditionally, economists have relied on simplified models or categorized jobs into broad groups. But LABOR-LLM takes a different approach, treating career paths like sentences, with each job being a “word.” By converting survey data from sources like the PSID and NLSY into resume-like text, researchers fine-tuned LLMs to predict the next “word” in a career sequence. The results? LABOR-LLM surpasses previous benchmarks, including the specialized CAREER model. This improvement comes from several factors, including the LLM's prior understanding of language and the rich context provided by full career histories. Interestingly, the model even recognizes the nuances of job titles, performing better with descriptive titles than generic numeric codes. Furthermore, the study highlights the importance of demographic factors like gender, ethnicity, and region, revealing complex interactions that traditional models might miss. While fine-tuning large models can be computationally expensive, the ability to leverage existing LLMs offers a powerful new tool for economists and social scientists. It opens doors for more accurate predictions and allows researchers to explore the dynamic nature of careers in unprecedented detail. What does this mean for the future? Imagine personalized career guidance, improved labor market analysis, and a better understanding of how our skills and experiences shape our career paths. While challenges remain, LABOR-LLM represents a significant step toward harnessing the power of AI to understand and predict the complexities of the modern labor market.