But the article precisely addresses that:
For example, in HN discussions people will often recommend Fastmail or Protonmail, but they've had their problems too (FM: 2017, 2020, 2022, PM: 2018, 2019, 2021). Especially given that these are much smaller services I'm not convinced that the risk is lower there. Any system is going to have to handle this sort of problem, and you're not going to find one that never has false positives.
What I'd like to know is : do FM/PM more easily enable to talk to a human ?