Devil's advocate: I think the argument is that the damage of the disease is disproportional to the people exposed, because the increased pairs of interactions are what can make it spread fast.

So doubling the number of people congregated in Rio won't just double the danger, but quadruple it. If car accidents obeyed a similar dynamic -- where they increase (much) more than proportionally to the number of vehicle-miles driven, we should be similarly more cautious about additional drivers.