GPG's vulnerability to quantum cryptography

Robert J. Hansen rjh at sixdemonbag.org
Tue May 13 20:00:34 CEST 2014


> What are your thoughts on these issues? Why do you keep using GPG, knowing
> that your data may easily end up out in the open on Google or The Pirate
> Bay a few decades from now?

Bluntly, my thoughts are that 99% of the people who talk about quantum  
computation couldn't identify a Hadamard transformation if they  
tripped over its brakets.

Shor's Algorithm requires 2N qubits, where N is the size in bits of  
the composite you wish to factor.  So for a 2048-bit certificate that  
requires 4096 qubits, representing a state space of over 10^1100.   
That's a quantum computer so ludicrously powerful that if one were to  
exist it would transform the world in ways we literally cannot  
imagine.  This is a quantum computer so powerful that it defies even  
the dreams of science fiction authors.

I literally lack the skill in the English language to describe just  
how eye-popping this thing is.  The best analogy I can think of is  
that we're a bunch of primitive hominids just beginning to learn how  
to knap obsidian into knife blades, and you're saying "What are your  
thoughts on how obsolete these knives will be once we develop  
thermonuclear bombs?  I mean, they're going to make these knife blades  
just ... *obsolete*."

> What are your thoughts on these issues? Why do you keep using GPG, knowing
> that your data may easily end up out in the open on Google or The Pirate
> Bay a few decades from now?

If that happens, I'll have much bigger things to worry about.  I'll  
let you worry about the thermonuclear age: for now, I'd rather focus  
on the advent of the Bronze Age.




More information about the Gnupg-users mailing list