Outgroups are from taxa that have been chosen apriori to analysis because they are well know to be outside of the group of interest. One cannot infer the root of the tree without an outgroup to anchor the root. So calling this clade an outgroup would be circular and arbitrary. Indeed, if the support for the apparent clade is small than you really don't know the topology of the tree either. That is fine but unrooting the tree altogether would give you an illustration of what you do know. Present it as an unrooted tree until you have independent data to define an outgroup.