Is it even possible to distinguish between such deepfakes and real footage at this point? Maybe they do that also so that there is no defensive line over claimi

Is it even possible to distinguish between such deepfakes and real footage at this point? Maybe they do that also so that there is no defensive line over claiming CSAM videos found are deepfakes and thus not a crime? Moreover even AI generated CSAM has to trained somewhere. It is not certain at all that an increase in AI generated CSAM market would necessarily mean less child abuse.