The existence of a consensus opinion isn’t an argument from authority. To claim so is an interesting form of the post hoc ergo propter hoc fallacy ('after this, therefore because of this'). The argument for agreeing with the consensus isn’t the existence of the consensus, but rather the same argument that caused the consensus to exist, the underlying body of research.

Framing it as a result that flies in the face of consensus is fun and exciting, especially since people who are science-literate know that a single compelling result can overturn a consensus formed by a large body of previous work. But framing it in that sense is unhelpful, since the vast majority of publications that “fly in the face of consensus” do not. It’s a lot less exciting to view this as one more paper on a large pile supporting MOND, that is still small in comparison to the pile that support dark matter. The paper is obviously worthy of attention and isn’t being dismissed out of hand, it is published in The Astrophysical Journal. But would Hacker News be discussing it if not for the contrarianism embodied by the former framing, as opposed to the latter?