If I claim that if you burn $1M it'll rain for a year, it's falsifiable, but we don't really need to test it. The burden would be on me to demonstrate scientifically how this hypothesis has any merit before it's reasonable to test it.