The increasing demand for real-time data processing has highlighted challenges in latency, energy efficiency, and computational resources. Traditional cloud computing faces issues such as bandwidth overload, slow response times, and concerns over security and privacy. In response, edge computing has emerged as a paradigm that performs computations closer to data sources, featuring localized, small-scale data storage and processing. This study proposes a novel methodology for sentiment analysis using the IMDB movie review dataset, comparing Edge AI and Cloud AI. We evaluated key metrics like accuracy, latency, and energy efficiency-to reveal the strengths and trade-offs of each approach. Our experiments show that Edge AI achieves superior performance: it reduces latency, improves energy efficiency, and achieves an accuracy of 89.48%, slightly surpassing Cloud AI at 88.28%. Notably, Edge AI exhibits significantly lower latency and energy consumption per inference. Analysis across edge devices-Raspberry Pi, NVIDIA Jetson, and Google Coral-shows that Google Coral achieves the highest accuracy with the lowest latency and energy consumption among tested devices.