Based on early solar system abundances of short-lived radionuclides (SRs), such as Al-26 (T-1/2 = 0.74 Myr) and Fe-60 (T-1/2 1.5 Myr), it is often asserted that the Sun was born in a large stellar cluster, where a massive star contaminated the protoplanetary disk with freshly nucleosynthesized isotopes from its supernova (SN) explosion. To account for the inferred initial solar system abundances of short-lived radionuclides, this supernova had to be close (similar to 0.3 pc) to the young (similar to 1 Myr) protoplanetary disk. Here we show that massive star evolution timescales are too long, compared to typical timescales of star formation in embedded clusters, for them to explode as supernovae within the lifetimes of nearby disks. This is especially true in an Orion Nebular Cluster ( ONC) type of setting, where the most massive star will explode as a supernova similar to 5 Myr after the onset of star formation, when nearby disks will have already suffered substantial photoevaporation and/or formed large planetesimals. We quantify the probability for any protoplanetary disk to receive SRs from a nearby supernova at the level observed in the early solar system. Key constraints on our estimate are: ( 1) SRs have to be injected into a newly formed (<= 1 Myr) disk, ( 2) the disk has to survive UV photoevaporation, and ( 3) the protoplanetary disk must be situated in an enrichment zone permitting SR injection at the solar system level without disk disruption. The probability of protoplanetary disk contamination by a supernova ejecta is, in the most favorable case, 3 x 10(-3). We propose instead that Fe-60 (and possibly Al-26) was inherited from the interstellar medium.