I find "absolute minimum" an unfortunate choice of phrasing when the process necessarily has large elements of randomness and could in principle well proceed from a single cell getting successfully infected.

But it's interesting to note that the way HIV generates its infamous genetic variability can result in a remarkably large fraction of the virions being mutated beyond viability.

https://journals.plos.org/plosbiology/article?id=10.1371/jou...

The infection risk would have to involve a factor of virions getting to a susceptible cell without getting destroyed anyway of course, but given that many of the contenders might not be capable of infecting even if they get there it seems even more understandable if one would usually end up with 10 as the number needed to satisfy some chosen level of statistical significance.