FWIW Grover's algorithm allows a quantum computer to do brute-force searches in sqrt(N) time, reducing SHA256 to the same difficulty as brute-forcing a 128-bit hash. This won't be practical any time soon (the constant time factors for quantum computers are huge and will plausibly remain so for some time), but in terms of computation time complexity (and not in any other sense), this is a "more efficient" way to mine Bitcoin than the way it's currently done.
Even if AI solves scalabile quantum computers and makes breaking hashes with Grover's algorithm practical, I think that shouldn't count as "creating a more efficient algorithm", since the algorithm itself exists already.
@BrunoParga it is quite literally playing guess-the-hash. AI will eventually find breaks in some of these, but Bitcoin is based on SHA-256 which is extremely resilient
@barbarous Yeah this sounds like a question posed by someone who doesn't know how Bitcoin mining works.