It's a bit ambiguous as to how to read that. I'm going to assume that you mean that you're setting a non-zero prior probability, which is reasonable enough. However, that's currently multiplied by zero evidence as far as I've seen.

Otherwise confirmation bias will simply take over and lead us astray. I mean, it's not exactly falsifiable. I could just as well say that most of the people I disagree with are actually just Perl scripts, not real people.