The general audience verbiage still changes over the time axis, though. Are you going to have different training sets for different time periods, and were would you cut the threshold between different corpuses? There's a lot of continuous flux, as well as discrete moments when a phrase turns sour when it was positive or neutral in the past. However, these are still classification issues which NNs could potentially resolve, but are still issues.

But it's certainly true that texts written to be specifically & clearly informative will always be easier to translate than fiction, opinion pieces, or informal conversations. It all depends on what you consider to be "high" on the result spectrum.

The successes already achieved leave the focus on the existing shortcomings as the ceiling, not necessarily on adding refinement to what it already does respectably well.