> Is prediction differs from information about future?
Yes, prediction doesn't and can't contain information (in the technical sense) from the future, since by definition it's based only on what we know in the present.
> Do we have some ideas why rules of physics will be different in the future?
No. It's perfectly reasonable to assume that the rules of physics will be the same in the future. Just as it's perfectly reasonable to assume that they're the same inside a black hole. But in both cases we can't know.
> prediction doesn't and can't contain information (in the technical sense) from the future, since by definition it's based only on what we know in the present
In the technical sense the future _is_ driven purely by the information available in the present. Itβs a function of t-1.
The only reason future should be different from prediction is if the set of information available to the predictor about the present is different from the actual set of information about the present.
Isn't this presupposing some hidden variable theory or superdeterminism to be true? If quantum interactions are truly random, the future is driven both by the information available in the present and randomness.
"Reasonable" is an unfortunate word IMO. An assumption doesn't come out of reasoning. We can't even estimate the probability of this assumption being right because we can't see the future.
The best way this word fits is that we can reason towards the conclusion that whether laws will change or not is an assumption.
Yes, prediction doesn't and can't contain information (in the technical sense) from the future, since by definition it's based only on what we know in the present.
> Do we have some ideas why rules of physics will be different in the future?
No. It's perfectly reasonable to assume that the rules of physics will be the same in the future. Just as it's perfectly reasonable to assume that they're the same inside a black hole. But in both cases we can't know.