Based on observations of other people using it, "AI" is already "poisoned" with too much low quality training material I'm not interested in searching The idea

Based on observations of other people using it, "AI" is already "poisoned" with too much low quality training material I'm not interested in searching The idea of intentionally "poisioning" an "AI" model seems to be the logical evolution of so-called "user-generated content" (UGC). It makes sense that non-cooperation is the eventual outcome of exploiting internet subscribers through coerced intermediation. For years now companies in Silicon Valley and Redmond, and other so-called "tech" companies elsewhere, have been generating revenue by using free information and UGC as bait for surveillance ad targets but without the licensing or production costs