Believe is something religious people do. The great thing about science is belief is unnecessary. If someone makes a fantastic claim it is reasonable to be skeptical. Skepticism is not an obstacle when you can prove that your claim is true.
Unfortunately, we are not all equipped to replicate all experiments, so even scientists rely on a significant amount of belief and faith.
This is trivial. If this superconductor is real then people will replicate it. People will publish results and once we have multiple independent verifications we can trust that it's real.
If it is also practically applicable then it will start to be used in real appliances and when we see it working we know it's real.
Science is frequently wrong. False positives, fabricated results etc are published on a daily basis. We should not accept one study's results as fact, we should not blindly believe in it, that's not how science works. We should understand that there are degrees of trust and certainty. A qualitative study is highly unreliable, a quantitative study's reliability depends on specific details such as sample size, methodology etc.
All a study says is "we did this and saw these results". We don't know that it isn't fabricated, we don't know that the description is accurate, we have to take it with a grain of salt. But at some point we have enough data to trust claims like "this medication is effective at treating this illness" or "this material is superconductive at room temperature".
I don't mean the system's broken in any way, far from it, this is a healthy functional system of scientific discovery! I just think it's fascinating how much is predicated on good faith.
There's also a lot of things that do get disproved, but nobody ever hears about that. Journals generally publish positive results, they don't publish negative results or failed replications. The whole for-profit scientific journal system seems pretty messed up to me.
But this is what we have. So I think it's important that people understand the limitations of modern science. In many cases the science is solid - we know MRI machines work, they're used all over the world. So clearly the science behind them is solid. It gets more fuzzy when people publish studies like "People who eat dark chocolate get cancer less frequently" because I can guarantee you any study in that vein will have something like a 4% lower incidence of cancer in the chocolate group than the control but it'll technically be just above the arbitrary limit we use to determine statistical significance.
That's the kind of science I consider pointless bullshit. It's not enough to be statistically significant, it has to be solid. It has to show a clear and indisputable effect, if it doesn't then the study is pointless. "It is possible that eating dark chocolate has some positive effect on your odds of getting cancer" is a completely useless result. We knew that before you started the fucking study.
Beliefs of course vary in the level of justification (a) that can be supplied (b) that actually is present.