What sets science apart from most other methods of seeking answers is its focus on disproof. Your goal as a scientist is to devise experiments that can disprove a claim about the natural world.
This misconception rears its head most prominently in discussions at the intersection between science and public policy. Climate change. How to handle a pandemic. Evolution. Abortion. But I've even talked to scientists themselves who from time to time get confused about what science can and can't do.
The problem with believing that science proves things is that it blinds its adherents to new evidence paving the way to better explanations. It also leads to the absurd conclusion that a scientific question can ever really be "settled."
Any hypothesis that I invent at this very moment, is from this perspective in the best position a hypothesis can ever be. There is no disproof. There is even no coherent argument against it, because I literally just made it up this second, so no one had enough time to think about it and notice even the obvious flaws. This is the best moment for a hypothesis... and it can only get worse.
I understand that there is always a chance that the new hypothesis could be correct. Whether for good reasons, or even completely accidentally. (Thousand monkeys with typewriters could come up with the correct Theory of Everything.) Yes, it is possible. But...
Imagine that there are two competing hypotheses, let's call them H1 and H2.
Hypothesis H1 was, hundred years ago, just one of many competing options. But when experiment after experiment was done, the competing hypotheses were disproved, and only this one remained. For the following few decades, new experiments were designed specifically with the goal of finding a flaw in H1, but the experimental results were always as H1 has predicted them.
Hypothesis H2 is something I just made up at this very moment. There was not enough time for anyone to even consider it.
A strawman zealot of simplified Popperism could argue that a true scientist should see H1 and H2 as perfectly equal. Neither was disproved yet; and that is all that a true scientist is allowed to say. Maybe later, if one of them is disproved in a proper scientific experiment, the scientist is allowed to praise the remaining one as the only one that wasn't disproved yet. To express any other opinion would be a mockery of science.
Of course, there always is a microscopic chance that H1 might get disproved tomorrow, and that H2 might resist the attempts at falsification. But until that happens, treating both hypotheses as equal is definitely NOT how actual science works. And it is good that it does not.
In actual science, there is something positive you are allowed to say about H1. Something that would make the strawman zealot of simplified Popperism (e.g. an average teenager debating philosophy of science online) scream about "no proof ever, only disproof". The H1 is definitely not an absolute certainty. But there is something admirable about having faced many attempts at falsification, and surviving them.
But, I wonder if you can describe H1 as being a stronger hypothesis than H2 by virtue of withstanding more and higher quality attempts to disprove it?
I think that when people are essentially honest and trying to find out truth, they can agree on reasonable rules. But there is no way to make the rules simultaneously philosophically satisfactory and bulletproof against people who are willing to lie and twist the rules in their favor.
For example, in real life you usually cannot convince crackpots about being wrong, but that is okay because at some moment everyone just ignores them. If you try to translate this into a philosophical principle, you end up with something like "argument by majority" or "argument by authority". And then you can have Soviet Union where scientific progress is often suppressed using these principles. But what is the alternative? No one can ever be ignored unless you disprove their hypotheses according to some high standard? Then the scientific institutions would run out of money as they would examine, using the high standard, the 1000th hypothesis of the 1000th crackpot.