We present ab initio molecular dynamics simulations of magmatic brines. Our study examines the structural and transport properties of highly concentrated brines at temperatures up to 873 K and pressures below 1.5 kbar. We investigate the binary system of NaCl and H2O to analyze local environments at varying salt concentrations. Additionally, we compute the electrical conductivity, which is crucial for interpreting magnetotelluric maps of volcanic areas. We explore the relationship between transport properties and the average number of free ions in the liquid. Furthermore, we examine the spectroscopic signatures of ion dynamics. To gain a deeper understanding of ion association dynamics, we employ machine learning potentials to simulate larger systems and enhance the statistical significance of our predictions.