I've noticed a very sudden uptick in users of FOSS software being so low trust that they will see a small change, assume it's much larger, and then retreat to some rationalization that it's still bad when shown it's pretty small (slippery slope / boiling frog type arguments). I'm not too familiar with this story in particular but I have been following the Systemd birthdate field controversy, and it's exhausting. I don't even think of myself as that high trust compared to the people taking issue, but it's like they're in a completely different world. Is this actually a trend (in specific, not the general loss of institutional trust) or am I only now paying attention?

Well, we can certainly disagree about the harms (fingerprinting) and slippery slopeness (AV started for porn, but we all knew it wasn't going to stop there)

But to get to the meat of your question, trust is lost through betrayal. Organizations have been deciding unpopular policies without consulting their users, or having meaningful methods for users to opt out or push back. For a long time, users assumed open source would be the last bastion of privacy and user freedom, and then were shocked when those values were not actually shared by maintainers.

The paternalistic perspective that the organization knows users hate something but push it anyway is always going to lose trust. That practice needs to stop, and instead consider how open source can treat the user base as important stakeholders.