"Appeal to consequences" makes it sound like the Thermian argument is a logical fallacy that establishes an incorrect way of forming a belief. The idea is that since our beliefs about what's true in the world should not be shaped by the social consequences of that truth, appealing to consequences is faulty logic.
Right, so far so good.
In science, where facts can be established through rigorous methods, that might be true. But I think social truths are determined by their consequences simply because social facts are man-made — and thus, we should probably keep the ethicality of their consequences in mind.
The problem with this is that it requires weighing every conceivable interpretation, which is not only impractical but also impossible, and then weighing contradictory interpretations against each other to decide which should be given more weight. Hence why the ethicality should be viewed as inherent to the nature of the action unto itself, rather than the consequences that they generate. For instance, murder is inherently wrong unto itself, not because the consequences might be deleterious for people (which correctly notes that it's not okay to murder someone even if they're a "bad person" or if you think no one will care). Of course, this also points out that fictional instances of a bad thing aren't the same as actual instances of it, since RPGs tend to involve murder on a large scale without anyone being hurt by it.
Does the Thermian argument say that the socially undesirable consequences of a fictional fact weigh in on whether we take that fact to be acceptable? Yes it does. But given how the fictional worlds we consume have an effect on our worldviews whether we like them or not, I don't think this consideration is fallacious. Yes, older works of fiction often have species created in the image of their creator, but knowing what we do on critical theory and how such arguments have been used to demonise or oppress real peoples in the past, our fictional worlds are better off not having such species.
The idea that what we consume has an effect on our worldviews is an idea which remains controversial, largely because the exact nature of what that effect is remains difficult (at best) to measure, let alone attribute to any particular source. This is especially true with regard to self-evident fiction (as opposed to, say, media that presents itself as factual in nature), which adults (presuming they don't have mental health problems preventing them from distinguishing between fantasy and reality) will be necessarily predisposed to dismissing in terms of shaping a worldview. For an example of this, notice how (at least until recently) violent crime rates were going down despite a rise in violent video game availability.
Which isn't to say that there's no effect at all, but in most cases this results in very short-term changes in mood in response to stimuli, such as sad movies making us feel sad, action films making us feel excited, porn making us aroused, etc. But those are all short-term effects, and are highly obtrusive in nature (e.g. there's no subconscious "programming" going on). To that end, the idea that repeated consumption of fiction will cause people to internalize (and translate, since quite often that fiction doesn't have a 1:1 parallel between what it presents and the real world) the messages it contains, which as noted often need to be translated anyway, is a much harder row to hoe. While I have no doubt that people have
strongly held feelings about what they find personally off-putting in an RPG, to bring this back around to the context under discussion, I don't think that's necessarily a compelling argument unto itself for changing the presentation of the material in question.
