scam alert false positives

i started a discussion about scam alert false positives. i provided some examples of said false alerts.

the discussion was deleted. possibly because i offered the examples about the false positives.

so. how do i explain the problem without being able to explain the problem?

also. i have now been flagged for "spam" because i needed to explain why that is a problem. let me use this time to explain in a roundabout way what the problem is without giving any examples whatsoever.

here is my attempt at "explaining without explaining": i am italian. Italian is an heavily gendered language. Greetings are often abbreviated with a period, but after said period one shows the last two letters to show respect to the person being greeted by (essentially) clarifying the gender of the object of the discussion via the gendered part of the word.

(unfortunately it is just male or female, no "neutral" or "other" option is present because the italian language was codified in the present state almost 2 centuries ago and one would have to rewrite it from scratch to add anything that is not male or female, as we use that for animals and things too. yes. it makes no sense. but such is life.)

therefore all greetings take the form of what gets recognized as "a link" (something, then a period, then two letters) by the scam alert.

Meanwhile actual scammers that try to use non-latin characters that look like latin letters to pass themselves as banks or... whatever... they never get caught...

...because they use passthrough links that are legitimate.

So. As of today i turned off the scam alert feature for giving me only false positives and quite frequent ones, while never alerting me of the few scams that pass through the spam filter.