Quantum Machine Learning Algorithms: Read the Fine Print says:
By using a quantum computer, one could dramatically accelerate the simulation of quantum physics and chemistry (the original application advocated by Richard Feynman in the 1980s); break almost all of the public-key cryptography currently used on the Internet (for example, by quickly factoring large numbers, with the famous Shor’s algorithm); and maybe achieve a modest speedup for solving optimization problems in the infamous “NP-hard” class (but no one is sure about the last one).
Why is no one sure about the last point? Don't quantum computers achieve polynomial speed-ups for Circuit-SAT, an NP-hard problem that some optimization problems can be translated to?