Machine perfusion (MP) preservation is potentially one of the most significant improvements in the field of liver transplantation in the last 20 years, and it has been considered a promising strategy for improved preservation and ex situ evaluation of extended criteria donor (ECD) organs. However, MP preservation adds significant cost and logistical considerations to liver transplantation. MP protocols are mainly classified according to the perfusion temperature with hypothermic machine perfusion (HMP) and normothermic machine perfusion (NMP) being the two categories most studied so far. After extensive preclinical work, MP entered the clinical setting, and there are now several studies that demonstrated feasibility and safety. However, because of the limited quality of clinical trials, there is no compelling evidence of superiority in preservation quality, and liver MP is still considered experimental in most countries. MP preservation is moving to a more mature phase, where ongoing and future studies will bring new evidence in order to confirm their superiority in terms of clinical outcomes, organ utilization, and cost-effectiveness. Here, we present an overview of all preclinical MP studies using discarded human livers and liver MP clinical trials, and discuss their results. We describe the different perfusion protocols, pitfalls in MP study design, and provide future perspectives. Recent trials in liver MP have revealed unique challenges beyond those seen in most clinical studies. Randomized trials, correct trial design, and interpretation of data are essential to generate the data necessary to prove if MP will be the new gold standard method of liver preservation.