I didn't say it was. Note I made a big distinction between the two kinds of problems. One is a problem of player expectation contrasted with what the game is giving you, the latter an problem of designer expectation with what the players at the other end are like. I might have opinions about what sorts of design are liable to into problems more often (which contrary to some people in this discussion I don't think is a trivial issue) but there's a big difference between the two.
OK, you are drawing a cognizable distinction, at least in this case, but I don't think it matters that much. I can reformulate my observation to avoid said distinction, and the same questions arise. Why is it that D&D's PC/GM fiction ownership and direction paradigm privileged? Surely THAT is of the same 'kind' as PbtA-style game's fiction ownership paradigms, yet one is labeled as dangerous, the other is benign in your telling. I'm perfectly willing to accept evidence that somehow playing an RPG where the players get to participate in the fiction (or other models that aren't exactly D&D-like in some degree) is 'disruptive', but I am very suspicious of simply accepting it as 'obvious' or any such thing. Many people think their opinions are 'obvious facts' or 'common sense' merely because they represent a lens through which they see the world.
Which is what I'm talking about in the second category.
I wasn't talking about either category at this point, I was simply agreeing that some types of games, and probably D&D-like games (in some sense) might be in this category, are very common and that publishers might reasonably assume that it is likely their audience for new game X is familiar with that paradigm and take more effort to carefully draw the distinctions between it and X. This is regardless of where in the game these distinctions fall, though I again don't disagree with you that play process and authority matters are more fundamental than 'how hit points work' and such (but very often the later kinds of things are cited as burning issues, so they are not without import, far from it).
I think the first is far more critical, though, and I think, if anything, game designers are more likely to make assumptions about it than the second. I think this is likely because just seeing other designs will tell you a lot of people seem to carry certain expectations or wants different from what you're giving them, and its probably doing nobody any favors to not let them know that.
I would simply state that I think if you write a 'D&D-like game', again in some general sense, you would be well advised to also tell people what the roles and process are in that game. Actually, one of the big problems I have with that general category of games is how UNCONSCIOUS they are of their strong assumptions and whatnot. There are all these potent social elements inhered into them without any analysis or explication whatsoever. So, I actually want to turn the tables on you and suggest that 'traditional' (not that I like the word) RPGs desperately need to be more self-reflective. Many of them never really cleanly state outright what they expect, whereas ANY PbtA or other story-focused type of game will almost invariably lay it all right out on the table from page 1! Read Dungeon World, you can get a free download of the PDF, you will instantly see that. You will be hard-pressed to find an 'indie game' which doesn't do the same. Heck, they have to or else they will be mangled in play!
But going in with different expectations about how players interact with each other, and with GMs is both easier to do, and far more problematic to fail to explain.
Again, I don't see why this is something that only one group of games is obliged to look at, and again I don't believe the type of games you are referring to generally fail to do this.
(There is a weird in-between case you can argue about power relationships between GMs and players, but I think what I'm talking about is more related to whether a GM is expected to have an adversarial, supporting, or somewhere in between relationship, That's only liable to disappear if there's no meaningful GM role at all, far as I can tell.)
See, in the end, I don't think the roles are really different at any table that ACTUALLY WORKS. It may be that in some games, rarely but I've experienced it, you really just buy the popcorn and sit back and acquiesce to get fed a story. This is rare however, but because the REAL expectations, or maybe we should say REQUIREMENTS, to make a game work are not even known by the game authors of most 'trad' games all these weird semi-unworkable arrangements arise.
Honestly, ONE WAY, not the way I normally do, but a way, to explain D&D 4e, is as a traditional game that is acutely self-aware and goes some way to describe this relationship. In other words, you CAN play it as a standard variation of D&D, and not as a story game (though I know
@pemerton will point out certain areas where you cannot square it with that, its a weird beast of a game). This was my initial impression of the game and reaction to it was "interesting, this game is very self-aware, unlike D&D generally." You see that in the way the rules are designed for good game play as a prime consideration. You can also see it in the various 'structure of play' areas, and 4e actually spent a good bit of time on those. I believe THIS is actually the burning heart of what was objected to about the game. GMs saw within it a kind of a mirror reflecting back at them a view of both how they are actually playing, and of what the flaws are in the generally assumed D&D (IE as 5e would have it, or 3.x, or 2e) play structure. I guess they didn't like what they saw in a lot of cases!