Henryk Plötz on Nostr: Also, I mean, it's *trivially* countered by AI scrapers. It needs exactly one prompt ...

Also, I mean, it's *trivially* countered by AI scrapers. It needs exactly one prompt "Hey Claude, this weird font replaces words with other words, look into the font file to find which and undo it".

Normal screenreaders on the other hand are royally fucked.

In security we say that something is secure if the attack is (much) harder than legitimate usage. Every once in a while you see something marketed as 'security' where that is reversed. This is one such case.