Just because something can't be proved (or disproved), with certainty, doesn't mean it isn't true. Thus, we must embrace uncertainty as epistemic and embrace uncertainty whole-heartedly in science with Bayesian inductivism as a replacement for falsifiability.
Hiesenberg's uncertainty principle makes this clear, the principle regards uncertainty in measurement not indeterminateness and fundamentally limits the power of falsifiability -- without certainty, you can not disprove. This uncertainty is due to the wavelength of light and light's impartment of physical momentum onto a particle (thus changing it) during measurement. This has nothing to do with how a particle behaves without measurements. I believe they ride some kind of unmeasureable, but believable and describable, pilot wave.
Thus, Bayesianism is the best anyone can do to explain anything -- or, certainty is unknowable or unobtainable. And so, science is a collection of inductive degrees of belief that are updated in light of new evidence.