judgments. But privacy advocates might worry about whether such systems occasionally make mistakes
This is the BS part of these arguments. They shouldn't be judged on if there will ever be mistakes made but rather will they reduce mistakes made by humans currently?
