Laser Powder Bed Fusion (LPBF) is widely considered a key enabling technology of the future. In order to realize its full potential, however, reproducibility and in-process quality inspection capabilities have to meet industrial requirements. The application of novel sensor technologies such as hyperspectral cameras and intelligent data evaluation methods like machine learning models will allow more reliable manufacturing processes. This article discusses the value of snapshot hyperspectral imaging as a means to predict process states and defects with the help of machine learning algorithms. The imaging technology is presented and its characteristic advantage of providing spectral as well as spatial resolution is weighed against the drawbacks of low temporal resolution and reduced spatial resolution. Besides, different methods and configurations for in- process data acquisition and subsequent data labeling are explained. Finally, the utilization of this monitoring approach to the LPBF-processing of magnesium alloys is discussed and results are presented.