Internet users tend to interpret comments they don't like as "paid accounts", bots, shills, spies, astroturfers, foreign agents, brigaders, and sundry other sinister manipulators. This is one of the biggest psychological mechanisms on the internet, so commonplace that the site guidelines specifically ask people not to do it:
https://news.ycombinator.com/newsguidelines.html. No doubt there's something hard-wired in us that makes us jump from displeasure to suspicion.
That's not, of course, to say that abuse doesn't happen. It's simply to say that when considering abuse, we need to look for actual evidence. It's impossible to evaluate vague general claims. We need to see specific links. For example:
> I often check an account and notice it was not active for a substantial period of time before a sudden change in tone is noticed.
As the site guidelines ask, you should send us specific links so we can look into them. Such perceptions usually turn out toe be rife with assumptions and projections, but occasionally we find evidence of abuse, and when we do, we crack down hard on it. The vast majority of it that we've seen is small-scale commercial abuse, though—voting rings, commenting rings, and various types of spam.
When it comes to disagreement on hot political/social topics, we all need to get used to the fact that society at large (indeed the world at large, since HN is a highly international site) is a lot more divided than we would like, and stop resorting to cartoonish Boris-and-Natasha explanations to explain away the unwanted facts that (a) a lot of people sincerely and strongly disagree, and (b) everyone has good reasons for their views, even when they're wrong or we feel they are.
I wrote a long post about this here: https://news.ycombinator.com/item?id=27398725. Note that all these arguments are limited to HN. Larger sites have different problems to deal with.
Even if you assume that there are Sufficiently Smart Manipulators [1] planting carefully crafted comments in undetectable ways, the conclusion that leads to is pretty much the same as what the site guidelines call for anyway: we should respond to bad arguments and false claims by patiently and respectfully providing better arguments and true information. If a community can do that, then hostile pathogens aren't going to kill it—and if it can't do that, then it's going to self-destruct anyhow. In either case, hostile pathogens aren't a fundamental cause, and the solution is to focus on and strengthen our own responses. The only viable long-run answer to the Sufficiently Smart Manipulator is the Healthy-Enough Community.
[1] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...