We present the highest spatial resolution ALMA observations to date of the Class I protostar WL 17 in the rho Ophiuchus L1688 molecular cloud complex, which show that it has a 12 au hole in the center of its disk. We consider whether WL 17 is actually a Class II disk being extincted by foreground material, but find that such models do not provide a good fit to the broadband spectral energy distribution (SED) and also require such high extinction that it would presumably arise from dense material close to the source, such as a remnant envelope. Self-consistent models of a disk embedded in a rotating collapsing envelope can nicely reproduce both the ALMA 3 mm observations and the broadband SED of WL 17. This suggests that WL 17 is a disk in the early stages of its formation, and yet even at this young age the inner disk has been depleted. Although there are multiple pathways for such a hole to be created in a disk, if this hole was produced by the formation of planets it could place constraints on the timescale for the growth of planets in protoplanetary disks.