Laser powder bed fusion (LPBF) additive manufacturing is capable of producing Nitinol stents of complex shapes, but a slight change of Ni content during LPBF can result in the stents not being suitable. Thus, there is a need to predict how laser power (P) and thus energy (E) affects the two important outcomes of Nitinol LPBF: defect level and Ni content. In this study, how Ni content may change during LPBF has been studied under two conditions: a sufficiently low P (thus E) but close to fully dense condition and a sufficiently high P that has caused keyhole pore formation. The loss of Ni has been found to be 0.36at% and 0.56at% for the sufficiently low and sufficiently high P conditions. Thus, 0.4–0.5at%Ni loss is expected for defect-free Nitinol LPBF. The morphological features of the melt pool affecting defect formation and relating to how P affects Ni loss will be discussed. With the significant loss of Ni due to LPBF, for the starting powder, which contains 50.5at%Ni and is fully austenitic, the LPBF as-built structure has remained mainly austenitic, but the samples also contain a small amount of martensite.