Phase-change materials (PCMs) are an important family of alloys employed in non-volatile memories and neuromorphic devices. These devices exploit the ability of PCMs to undergo rapid and reversible transitions between crystalline and amorphous phases at moderately high temperature. At ambient conditions, both phases are very stable. These properties imply a very strong temperature dependence of the crystallization kinetics, which has been attributed to the high fragility of the supercooled liquid phase. The potential energy landscape (PEL) is a powerful tool for understanding the thermodynamics and dynamics of glass forming liquids. Exploration of the PEL can be carried out using molecular dynamics. However, a large set of configurations in a wide range of temperatures is needed for a good sampling of the PEL, requiring long simulation times beyond the reach of ab initio methods. In this work, we conduct massive molecular dynamics simulations based on a neural network potential to investigate the PEL of liquid GeTe, a prototypical PCM. Using thermodynamic integration, we compute the configurational entropy as a function of temperature in the deeply supercooled regime. We also calculate the viscosity and the relaxation times in the same temperature range. Finally, we use the Adam-Gibbs equation to extrapolate the viscosity down to the glass transition temperature and estimate the fragility of the liquid.