Reading through the original question and the subsequent comments there are a couple of things that spring to mind. As I'm still fairly new in the WikiTree game, please let me know if the below already exists or is done in some other way - or if it's been suggested and rejected before. A clear process does, however, appear to be lacking judging by the other responses here and in the previous discussions I've lurked in, and the underlying root cause seems to me to be ambiguity in the use of the tags and what they could/should mean.
The blurb and its overlap with {{Uncertain Existence}} which ties in with one or the other being used for living people as well as imaginary ones. The blurb gives "hope" that it "could be possible" that the relationships existed. It is ambiguous and non-specific, and thus a fair waste of space and effort. Unfortunately so is the label {{Uncertain Existence}} - there's nothing "uncertain" about it - if there's no solid proof or thorough research to provide a convincing amount of incidental evidence, then until such works become available, the profile should be marked with something that clearly states that the person is {{Not Proven to Exist}} (or similar). This is specific enough to point out there's no proof.
This leaves us in a quandary about people who did exist but who have no known or proven genealogical connections. As a much more modern example, the father of a profile I'm currently working on has been included on WikiTree without evidence. There are several other candidates for the part of father in this case, and it will become necessary to orphan the father as he has {{No Proven Ancestry or Descent}} (which could easily be treated also as {{No Proven Ancestry}} and {{No Proven Descent}}). The text in the biography should then address the alternatives and other interpretations. But wait! There's more!
One of the alternative fathers for the profile I'm looking at appears in secondary sources, some with fairly specific information. They're not referenced and I do not have access to the archives of the Diocese of Gothenburg for the period concerned. His existence has been called into question in other secondary works. I would say he is of {{Uncertain Existence}} and more research is needed.
With a clear and unambiguous set of easy-to use tags there should be no need for extensive apologetic blurbs. It should hopefully assist in making the process easier to understand, and will ensure that those who stumble upon WikiTree from Google are in no confusion over the veracity or accuracy of our work.
The second point that was raised below (I'm sure for the umpteenth time) was the use of authorities, and in specific MedLands.
MedLands is the work of a one-man-band, and based, ultimately, on Europäische Stammtafeln (hereafter ES) by Detlev Schwennike (now deceased). It's given additional authority by being associated with the Society For Medieval Genealogy, of which I'm a member. Their annual publications are worth the price of admission and hold a high, academic, standard. MedLands, however, does not. ES being the core, means that dubious relationships from therein are copied over into MedLands (albeit marked as not proven). Cawley cites sources in Latin and works much in the same way I would work myself, to be fair, however he is careless when working in languages other than his native English, and any primary sources he cites should be cross-referenced both in their original and in translation. MedLands is a great place to start and it's an invaluable source in locating the primary sources, but it is a source I would only reference in discussions about opinion or as a "See Also", never as a source of a fact. As ES is the base for MedLands, it should also be made clear that a lot of the material therein relies on old and outdated research, which a quick glance through the bibliographies reveal.
Both MedLands and ES suffer from the same problem - they're aiming at breadth in a subject matter that requires the researcher to go into depth. And that's where WikiTree is far superior: with all the varied interests here, it's quite possible for individual researchers to go into the depths and still make a valuable contribution to an overall whole.
That's my 5 units of your lowest-valued currency. :)