I mean, no-one sensible would _ever_ take _any_ chip manufacturer's claim about performance seriously without independent verification; it is very much, at the best of times, a space where people tend to be, ah, _optimistic_ about unreleased stuff.

I can't think of any chip-manufacturer who _hasn't_ made claims about their chips which turned out to be grossly exaggerated or outright false when exposed to independent analysis.