What we have is a giant GAN - google's AI trying to spot the real bad actors from not, against the bad actors AI trying to fool the google AI. I suspect your e

What we have is a giant GAN - google's AI trying to spot the real bad actors from not, against the bad actors AI trying to fool the google AI. I suspect your experience is a sign that the bad actors are getting the upper hand and one of the problems in this space is human intervention can't really scale against machines. It could be a fight for survival for these companies whose huge market cap is based on the enormous scaling benefits of automation - however if it turns out that the automation processes can't successfully scale once automated fraud is taken into account then the only solution might be to descale the operation - making them vulnerable to traditional companies.