Quantum computers that use Grover's algorithm in the case of breaking symmetric encryption algorithms such as AES (Advanced Encryption Standard) are only able to weaken their strength or reduce all possible encryption keys to half.
That's not accurate.
As I said above, Grover's algorithm allows a problem to be solved in the square root of n-time. Half of 2
256 is 2
255. Grover's algorithm reduces 2
256 to the square root of 2
256, which is 2
128.
But in Bitcoin using the 256-bit ECDSA digital signature system, the possibility of encryption keys that can be done effectively Brute Force from 2256 to 2128.
If you are talking about solving the ECDLP, rather than brute forcing part of a seed phrase as above, then you are now talking about Shor's algorithm, not Grover's. Shor's runs in polylogarithmic time, and can factor a
k bit number in
k3 time. A sufficiently powerful quantum computer (again, decades away) could easily break the ECDLP.
Making more copies is risky too.
The best solution is have a set up where compromise of one back up is insufficient to steal your coins, such as separate seed phrase and passphrase back ups, or a multi-sig. Make two copies of each part. That way you have redundancy against accidental loss as well as greater protection against theft.