3 ms·
And I don't get it. If the intent is "we do not want people to talk about death", then surely the filters / censors would cover all different spellings and euph
by Cthulhu_ 2y ago
And I don't get it. If the intent is "we do not want people to talk about death", then surely the filters / censors would cover all different spellings and euphemisms of it as well?
I'm 99% sure the organizations that do this - tiktok, roblox, etc - are well aware of the users' attempts of euphemizing around the filters, so they are OK with it. That, or it's from higher up but the legal whatsits can't keep up.
- StevenWaterman 2y agoIt's the advertisers that don't like discussion of death near their adverts. So services can say "we don't monetise content that mentions death" and then their algorithm downranks that content because it's unprofitable
- ryandrake 2y agoBut the people using euphemisms are also talking about death. Don't the advertisers object to that, too?
- adolph 2y agoThe advertiser's objections must be implemented with actual words since the semantic layer is not yet computable. Maybe an llm could associate "die" to "unalive" and treat them the same way for the purpose of advertiser demonetization; time will tell and as censorship tightens the circumlocutions will broaden.
- StevenWaterman 2y agoYes, but nobody really cares that much, either at the social media platform or the advertiser. They just want plausible deniability