Vice carefully flavors the notion of dystopia so that you’re pretty sure it is when the AI is biased. It portrays the problem not as one of not being able to f

Vice carefully flavors the notion of dystopia so that you’re pretty sure it is when the AI is biased. It portrays the problem not as one of not being able to fully use the model, but one of the model not being sufficiently tested as safe. Vice, and at least one of the interviewees, at least how he was portrayed, appear to be espousing the belief that open thought policing is the answer. Put differently, the problem with a closed model is that it could be hiding some dangerous information inside that could be unleashed on an unwitting user. The solution is to openly build a model that censors the bad ou… This is sounding dystopian, too…