It is worse than nothing, because incorrect positive results and non-published negative results lead people to waste years of their lives building off something which is not true.

Yep, and "nothing" (as in "the effect of X on Y is zero") is a meaningful result, possibly a very important one, but being "nothing", it has a lot lower chance of actually getting published.

The issue is “nothing” is not sufficient to assume zero effect, it just means the effect wasn’t measured. Actual progress means trying to replicate both positive and negative results.

Well yeah, but if you're trying to "cure cancer", and in a (just made-up) theory an acidic environment should help with cancer, so you try lemon juice, and it shows that lemon juice cures as much cancer as placebo does, that seems publishable to me.