We simulate economic predictions of professional forecasters using a set of large language models (LLMs). We construct synthetic forecaster personas using a unique hand-gathered dataset of participant characteristics from the Survey of Professional Forecasters. These personas are then provided with real-time macroeconomic data to generate simulated responses to the SPF survey. Our results suggest that AI-based forecasts align closely with the accuracy and distribution of human predictions, and perform especially well at medium- and long-term horizons, often outperforming human forecasts. Our framework offers a higher-frequency, lower-cost alternative to traditional survey methods. It also provides new insights into the behavioral and structural factors that shape professional economic predictions.