Ab initio molecular dynamics (MD) is entering in a new age related to the improvement of algorithms and increase in computer power. All this has allowed a dramatic jump in the size of the systems studied (hundreds of atoms at present time in standard applications), the scale of the simulation (10-50 ps), and, more importantly, the quality of the results. Nevertheless, even though AIMD is widely used when strong variations in electronic properties play a crucial role (e.g. chemical and enzyme reactions, catalysis, interactions at surfaces, surface scattering simulations, magnetism, spin polarized and charge transfer systems) there are still severe issues that await further developments in the field. On one hand, those related to approximate XC functional (e.g. underestimation of reaction barriers, localized states, calculation of van der Waals forces). On another, problems related to insufficient sampling (particularly relevant in simulations of rare events), the size of the systems and the use of AIMD for high-throughput calculation. New techniques are continuously being developed to overcome these limitations and it is expected that ab initio simulations will expand the range of applications in forthcoming years. This international meeting is intended to review the state of the art in the area and to clarify the dominant trends in the field for next years. We also expect from this meeting to join in an informal environment several of the most prominent researchers working in the area of ab initio molecular dynamic simulations with young emerging scientists.