I'm not entirely sure. But we have have mechanisms like Google Safe Browsing, and I can imagine that a similar mechanism could be used. Or there could even be a new classification in Safe Browsing and similar databases for newly registered domains that look like they are misleaing, and this could actually be fast enough if there was a 24-hour hold on new domains.