> We don’t even know whether we’re living in a simulation.

I think it's actually worse than this. This scenario suggests that our minds are capable of infallible reasoning yet we may not be able to trust our observations. Really, I don't think we can even trust our own mind, and therefore JTB is undefinable.

This strikes me somewhat similar to saying that we should not use Newtonian mechanics, because we know better that at subatomic level Newton's laws do not apply.

But Newtonian mechanics are still extremely valuable, worth discovering and understanding, and anyone catching a baseball is employing them quite proactively. JTBs are the Newtonian mechanics of epistemology. You can pick them apart at a deeper level and show how they don't really exist, but they are still incredibly useful.

I'm not saying that we should entirely disregard epistemology. After all, belief is necessary to make decisions, and thus to thrive. Although I think people should be mindful that knowledge may not be possible in actuality, just as modern physicists are mindful of relativity and quantum mechanics.

Also specific example with a cow is silly. Belief there is a cow in the field is not really justified. It just seems justified to the observer that can't tell difference between model and a real cow.

By what definition? The "notion" of "justified true belief" WAS the definition of knowledge prior to these complications. Just saying "everything is subjective" is more tedious than making meaningful distinctions, in my subjective opinion

I think GP means that the justification of JTB by definition is recursive and thus has no real foundation.

To call something 'true' is to know it is 'true'.

If we are bothering to debate whether knowledge is possible, JTB is unconvincing.