Carbon offset programs are critical in the fight against climate change. One emerging threat to the long-term stability of offset credits is wildfires, which can destroy forest offset projects and associated credits. However, analysis of carbon offset fire risk is challenging because existing models for forecasting long-term fire risk are limited in resolution and skill. Therefore, we propose an interpretable deep learning model trained on millions of global satellite wildfire observations. Validation results suggest substantial potential for high resolution, enhanced accuracy projections of global wildfire risk, and the model outperforms the U.S. National Center for Atmospheric Research's leading fire model. Applied to a collection of active U.S. forest carbon projects, we find that fire exposure is projected to increase 64% [48-96%] by 2080 under a mid-range scenario. Our results indicate the large wildfire carbon credit losses seen in the past decade in the U.S. are likely to become more frequent as forests become hotter and drier. In response, we hope the model can support wildfire managers, policymakers, and carbon market analysts to quantify and mitigate long-term risks to forest carbon offsets.