The worst thing about widespread fake information is the ease with which you can convince yourself that everything you don't like is fake. We don't really have a defense against it that allows for people with opposing perspectives to coexist.<p>I wonder if it's possible to somehow create a "consistency checker" that can check to see what claims are compatible with what other claims. Not a truth checker; that's far too dangerous and easily abused. But a system that just checks for consistency, combined with a "web of trust" where you can see the connections to people with different perspectives who you know are NOT liars/can connect to events you KNOW are true. That might help people trust information from outside of their comfort zone and identify problems with their own outlook.<p>There's precedent for that kind of a thing in the math world. MetaMath is an example of a system that verifies proofs based on substitution rules alone, plus whatever axioms you add to the system. Natural language news stories that need to correspond to reality are obviously a much different beast, and maybe it's impossible to do something equivalent with knowledge that isn't as rigid/formal as math, but it's tempting to try.<p>Could easily backfire and make the problem worse if done improperly, though. The risk is that such a system could be taken over/hijacked my ideologically motivated individuals to reinforce their own ideas and hide inconsistencies rather than shine a light on them.