This study introduces an advanced approach to automating the classification of field safety inspection findings using a fine-tuned Large Language Model (LLM). Achieving a high accuracy rate of 93%, the model dramatically reduces classification time, allowing organizations to quickly pinpoint critical safety issues and areas for improvement. Traditionally, manually categorizing thousands of inspection entries is a labor-intensive process that can take months. In contrast, the AI-powered system completes the same task in minutes, freeing up valuable resources for higher-value initiatives. The model systematically classifies findings into key safety categories—such as Emergency Preparedness, Safe Systems of Work, and Hazardous Material Management. Following rigorous testing and quality assurance, the solution is ready for deployment, enabling organizations to enhance safety standards, mitigate risks, and improve regulatory compliance. This automated classification system has the potential to revolutionize the analysis of safety inspections and can be applied across various industries where safety is a top priority.