At least an NGO will likely have a consistent point of view. The CN algorithm, apparently, requires “agreement from contributors who have a history of disagreeing.” Let’s say we have an entirely hypothetical scenario where the two primary political groups arguing over notes are a milquetoast centrist party and a far-right party susceptible to conspiracy theories; accordingly, any notes that are agreed upon will either be extremely obvious (“the sky is blue” but not, perhaps, “the president’s wife is not a man”) or will tilt center-right. That seems far from objective to me. And that’s to say nothing of thumbs-on-the-scale tweaks to the algorithm by the platform owner, which will be undetectable, or changes to the political makeup of the editors.
I don’t think there’s any way to algorithm your way out of non-trivial fact-checking. Tech is not the solution to these kinds of fundamentally social problems.
(I should add that the best-case scenario here is an emergent and stable cabal of intellectually-rigorous editors, perhaps of varying political persuasions, similar to what happened to Wikipedia. But that’s barely different from fact-checking by some NGO.)