For example, HN has a plethora of stories about abuses at and by major corps and they often stay on the front page. Someone with a more pro-business bias would look at the same front pages as you and see the opposite 'major trend'. In fact I'm sure someone is, right now.
As for shadowy brigades of astroturfers behind story flagging, that is so classic a cognitive bias that we just ask people not to go there. The users who flag stories you favor aren't shills—they're normal users like everyone else, who simply have different views about what belongs on HN. Exceptions to that exist, but they're so rare as to have zero explanatory power. If you want more info about this, I've posted countlessly about it: https://hn.algolia.com/?sort=byDate&prefix=true&page=0&dateR....
If an Uber executive or PR person were particularly unscrupulous (would that surprise anyone at all?), another method would be to say something really insensitive in a thread like this and start a flame war to trigger HN's ranking algorithm. Imagine how easy that would be... a somewhat carefully disguised comment with the sentiment of "depression is just weakness, suck it up!" would rightfully send this thread into a frenzy of angry commentary and downvoting.
Also, it would be an interesting read if you could publish some of the findings as a post or something.
I'm not saying it figures in any specific thread or context, just that it's possible that bot capabilities may outstrip HN's ability to detect them.
You may only discuss as a private person, stating your own opinion, OR disclose which company/government organization you represent.
People see comments that they don't like -> They send the link to their group of friends -> Their friends who share the same sentiment downvotes you in quick successions.