Yes, I agree. That was poorly phrased by me.
What I mean is that academia finds it structurally impossible to do research that ends with the conclusion that maybe there's enough research done into something for now, that perhaps there's no problem that requires this sort of academic attention, or perhaps we lack the tools to make useful predictions in a field at the moment. There are no feedback loops.
Imagine you go into a very small, very closed field like climatology or economics, do some research, and your conclusion is this: "climate is too complex for us to model with any certainty, our data sets are corrupted and low quality, we really have no idea what's going on and can't fix this any time soon". Or even "the climate is changing but not in any actually problematic way, there's nothing to do here".
This may well be a legitimate or correct conclusion (for any field of science), but it's also a career-terminating one. Research into the question of how effective research can be just doesn't get done by academia because there's no ground truth end goal - research isn't a means to an end, as in the private sector. In academia research is itself the end.
This is what I'm trying to get at. To leave climatology for a moment, look at how long it took for replication studies in psychology to start at any scale, and how much science - for decades - has been found to be completely bogus. It's a staggering amount. Every time I read something about the replication crisis I'm stunned by the enormous scale, and how much more there seems to be to uncover. Academia has simply not been funding "psychology skepticism" and to a large extent still isn't. The production of large amounts of nonsensical research is guaranteed by the incentive structure of academia, in which research is entirely self-justifying and in which it doesn't pay to shoot down colleagues in your own field.