You can go meta and think about what is knowledge and what is scientific method. Knowledge is inherently axiomatic. In other words there are certain unprovable

You can go meta and think about what is knowledge and what is scientific method. Knowledge is inherently axiomatic. In other words there are certain unprovable truths (“axioms”) you agree on as granted and then build your theories from based on certain principles (and this is “scientific method”). In this case “conservation of information” is one of the axioms quantum mechanics is build on. Now, thinkers operating on the fringe of our understanding and advancing our theories have a choice to make: (a) assume it’s true on the black hole event horizon - thus paradox and theories on how to resolve it, (b) assume it’s not true on event horizon. So we don’t know if it’s a paradox at all, we choose to think it’s a paradox because we want our theories to work everywhere. But it’s not necessary correct assumption. When these theories are discussed this is implicitly implied.