I think it is somewhat nuanced. In principle you are right, I would even go so far as to say: You cannot confirm a theory with experiment. But every test build

I think it is somewhat nuanced. In principle you are right, I would even go so far as to say: You cannot confirm a theory with experiment. But every test build to "confirm" a theory, is always also an experiment to falsify every other theory with a different prediction. Look at the Higgs discovery. The experiments were clearly designed to measure the Higgs and confirm the prediction. But you can state it as "they were designed to falsify the (at that time SM) theory that there is no Higgs". What is important to remember that a) you cannot absolutely confirm a theory (as in: rule out all others) with a finite series of experiments b) correlation doesn't mean causation c) causation doesn't mean correlation.