Yes, there is quite a bit of nonsense in pre-1500 profiles, especially those from early GEDCOM imports before Wikitree tightened the rules, and those of us encountering such profiles often have to make quite a lot of corrections. (A few months ago I spent a considerable amount of time correcting family lines for early medieval members of my family, and removing fictitious links to an illegitimate Plantagenet.)
But I think it would be unrealistic to suggest that family links should be included only where there is no dispute or room for doubt. Even for some monarchs and many leading aristocrats there are some question marks about parents and children, but there is still often enough evidence to give a probable set of family links, some certain, some a bit less certain. This is just the nature of the surviving evidence.
The surviving records themselves may be inaccurate. For instance there are fanciful fabricated deeds designed to bolster claims to land. The information in English Inquisitions is frequently based on people's recollections and statements at the time, which may not always be accurate or honest.
I think all genealogists and historians who do work on the medieval period accept the limitations of the evidence, and the consequent uncertainties. All we can do is seek to achieve the best we can with what is available, and be open to revising genealogies if new evidence comes along. The sort of proof one commonly finds for people of the last couple of centuries frequently just does not exist. Even the most authoritative genealogists - like Douglas Richardson - recognise that they may have come to some wrong conclusions and that their work will contain errors (as is explicitly acknowledged in the forewords to Richardson's books).
For post-1500 profiles too, evidence is not always 100% firm. As a simple example, only in the 1970s did a researcher disentangle two fairly prominent 16th-century members of the Nowell family whom almost all the experts had previously thought were one and the same person, but who were in fact contemporaneous cousins. And I have a large set of Ford ancestors who lived in the same part of Staffordshire in the 17th and 18th centuries and had a handful of first names, and for some of them it is hard to be absolutely certain which of them begat whom, and which of them burial records refer to.
The answer to all this is to give as much sourcing as possible; to use research notes etc to mention significant uncertainties; to make use of the Wikitree facility to attach a confidence level to a relationship; and, as I say, to be open to new evidence. Not to insist on a degree of certainty the pursuit of which is a chimera.
