No. The paper is about reducing a set, as far as I understand. It's not about mistakes. It's about the deliberacy of basically squeezing out the individuals who are vulnerable and who do not see the signs of being scammed.
One of them said, and I'll never forget, "Thank you for being reasonable".