Accurate crime prediction is crucial for effective law enforcement and security, enabling proactive resource allocation and risk reduction. Criminal behavior is influenced by complex, diverse socio-economic factors, necessitating advanced models capable of extracting intricate patterns from large datasets. This research presents a \textbf{methodological and applied comparison} of four primary categories of time series forecasting models: Statistical Models (\textenglish{AutoARIMA}), Machine Learning models (\textenglish{AutoLightGBM}), Deep Learning models (\textenglish{N-HiTS}), and Foundation Models (\textenglish{TimeGPT}). The study's \textbf{innovation} lies in (1) integrating these diverse categories in a single comparative framework tailored for security decision-makers, (2) explicitly applying cutting-edge AI, particularly \textbf{Foundation Models (\textenglish{TimeGPT})} with pre-training on vast, multi-domain time series, for crime prediction for the first time, and (3) demonstrating a comprehensive application using daily crime data from Chicago (2017–2019), with the final month serving as a challenging test set for assessing robustness against sudden fluctuations. Results indicate that Foundation (\textenglish{TimeGPT}) and Deep Learning (\textenglish{N-HiTS}) models outperform in accuracy, effectively capturing nonlinear relationships and complex seasonalities. Statistical (\textenglish{ARIMA}) and traditional ML (\textenglish{LightGBM}) models offer greater interpretability and faster training but are less adept at handling unexpected surges. This comparative, automated approach offers a \emph{practical solution} for security agencies seeking AI adoption without significant programming complexity. The research underscores time series modeling's role in enhancing security operations and explores new avenues for AI-driven proactive crime prevention using big data.