In one of my previous careers, I ran multi-million dollar research projects.
For some reason, the funders demanded accountability in regards to whether the outcomes we measured were from pure random statistical variation or whether we were measuring something real.
Some thoughts: Skepticism is a virtue in science. Signal vs noise measurement is tough to do well. Humans have to make a lot of hard decisions with imperfect and partial data. We do the best we can, but we shouldn't get too cocky about the certainty with which we do it.