The MEG experiment took data at the Paul Scherrer Institute in the years 2009-2013 and published the most stringent limit on the charged lepton flavor violating decay mu->egamma: BR(mu->egamma) <4.2 x 10^(-13) @90% C.L. The MEG detector has been upgraded in order to reach a sensitivity of 5 x 10^(-14), which corresponds to an improvement of one order of magnitude. The basic idea of MEG-II is to achieve the highest possible sensitivity by making the maximum use (7 x 10^(7) muons/second) of the available muon intensity at PSI with an improved detector, since MEG ran at a reduced intensity (3 x 10^(7) muons/second) in order to keep the background at a manageable level. The key features of the MEG-II are the increase of the rate capability of all detectors to enable running at the intensity frontier, and to increase the resolutions while maintaining the same detector concept. A new mass, single volume, high granularity tracker, together with a thinner muon stopping target, leads to better spatial, angular and energy positron resolution. A new highly segmented timing counter improves positron timing capabilities. The detector acceptance for positrons is increased by more than a factor 2 by diminishing the material between these two detectors. The liquid Xenon calorimeter has new smaller photosensors (VUV-sensitive SiPM) that replace current phototubes and improve in particular photon energy resolution. The results of the 2018 pre-engineering run, the first with all the sub-detectors, and the current schedule will presented. MEG-II, together with the next generation charged lepton flavor violation experiments Mu3e (mu -> e+e-e+) at PSI and Mu2e and COMET (mu -> e conversion) at Fermilab and JPARK respectively, will reach an unprecedent sensitivity in the next five years. On the same time scale accelerator upgrades are expected that will provide muon beams with intensities of the order of 10^(10)$ muons/second. At this extremely high beam rates, new detector concepts should be adopted in order to overcome the accidental background in mu->egamma searches. Some future directions will be discussed.