Looking for the invisible...
From the link:

The problem with all of this is you’re looking for something, a needle in a haystack,” Cushman says. “To build a detector, you need to make some assumptions about how shiny is the needle, and how big it is, and where the haystack says you should start looking first. So that’s what we do when we build these detectors. We think of the most likely place, and of course we could be wrong. And so as time moves on, and we don’t find the needle where we expected to find it, we look further and deeper into the haystack.