Knowledge conditions about the truth-value of most synthetic propositions (statements about "stuff") are quite challenging to attain. Analytic propositions are much easier, but devolve into tautologies. And justified true belief is not knowledge.

Much of what we think we "know" we instead believe. But it's also the case that the remaining uncertainty (in the face of massively convincing - if not 100% probative - evidence) is so minuscule that proceeding as if belief is actually knowledge isn't irrational. It's betting on very, very good odds.