Intel is launching an AI-powered tool that will detect hate and redact hate speech in real-time audio. It lets users (gamers, in this case) toggle the "racism" setting to "some," if you want to keep a little bit of racism for some reason vice.com/en/article/dyvgvk/i…
Intel’s Dystopian Anti-Harassment AI Lets Users Opt In for ‘Some’ Racism
Intel's 'Bleep' is an AI solution that will ignore a systemic problem rather than address it.vice.com
I’d guess that’s just determining how aggressive the AI is at censoring, right? Still though, who the heck thought that labelling made any sense in context? And if it is allowing through a percentage of racial slurs, who thought that was a use case anyone would want?