I am not sure why I haven't heard much chatter about this paper by Dr. Claus Peter Schnorr where he claims to have come up with a classical algorithm to break the RSA protocol. The construction is based on a so called "Shortest Vector Problem Algorithm". The actual technical content is far beyond my understanding but I have a few questions about the paper and its implications.

  1. Has the correctness of the algorithm been already proven or disproven?
  2. What is the scaling of the algorithm? If efficient (meaning polynomial in the number of bits), would this undermine the field of quantum computation(QC) as Shor's algorithm is held up as the central example of the usefulness of QC?
  3. I know that one can use post quantum cryptographic protocols to avoid Shor's algorithm. Maybe something similar can be done here (by increasing the number of bits if the scaling is poor). However, if it can even break current RSA protocols, can't one just store all the current passwords to decrypt it later(If my memory doesn't betray me, this was the reason behind the NSA investing millions into QC)? If so, what is your bank doing about it?