This study focuses on the integration of Large Language Models (LLMs) in the context of municipal evaluation reporting in Zimbabwe, with an emphasis on their ability to enhance report accuracy, pace, and analytic attributes. The study conducted a survey of 94 monitoring and evaluation (M&E) officers in municipalitie. The study identified significant barriers such as concerns of data privacy issues, inadequate digital abilities, and a lack of technology amenities. Despite such obstacles, the research identifies critical enablers in better report quality and efficiency, leadership support and training programmes as crucial for the adoption of LLM. Creating effective governance of data that would enable ethical and successful use of LLMs requires establishing an effective governance framework of data in the context of LLMs usage, continuous training, and increasing technological capacity building, and promoting a culture of innovation. This research indicates that LLMs have huge promise in improving municipal accountability and transparency, but it also highlights the need for region specific initiatives to meet developing areas’ unique needs. These results offer a valuable perspective on how AI can be applied in resource-limited settings, and they guide public administrators with practical recommendations on how to integrate AI in such settings.