Saturday, January 4, 2014

NSA classifies quantum computers

According to leaks reported by the Wash. Post, the NSA classifies quantum crypto research:
The National Security Agency is conducting what it calls "basic research" to determine whether it's possible to build a quantum computer that would be useful for breaking encryption. ...

“It seems improbable that the NSA could be that far ahead of the open world without anybody knowing it,” MIT Professor of Electrical Engineering and Computer Science Scott Aaronson told the Post.
We have to keep with all the others making no progress in this field:
Seth Lloyd, an MIT professor of quantum mechanical engineering, said the NSA’s focus is not misplaced. “The E.U. and Switzerland have made significant advances over the last decade and have caught up to the U.S. in quantum computing technology,” he said.
Here are the qubit details:
(S//REL) Level B QC - Classified theoretical and/or experimental research in the design, physical implementation, and operation of quantum computers, as established by the Laboratory for Physical Sciences/R3. The boundaries are based on the number and quality of qubits, realism and specificity of design, control precision, and detail of analysis. While these boundaries may change over time, as of the publication of this guide, the values are:

(1) (S//REL) Detailed engineering design of 51 or more physical qubits;
(2) (S//REL) Implementation and operation of a high-fidelity 21-or-more physical-qubit device; or
(3) (S//REL) Implementation and operation of three (3) or more logical qubits, with sufficient speed and precision to allow preservation of quantum information and logical gates between the qubits.
So they classify 51 physical qubits, but a mere 3 logical qubits.

A qubit is not really a qubit unless it can preserve quantum information and be operated on by logic gates. So the NSA is using the term "logical qubit" for a true qubit, which has never been demonstrated. Apparently it is customary for quantum computer researchers to claim some large number of qubits, but they are not real qubits. The NSA will have a major top-secret breakthru if it obtains a device with a mere 3 logical qubits.

I am still banned from Aaronson's blog. He recently posted, in response to skeptics:
In my personal opinion, proving that QC can work would be at least as interesting—purely for fundamental physics, setting aside all applications!—as the discovery of the Higgs boson was. And our civilization (OK, mostly the EU) decided that finding the Higgs boson was worth $11 billion. It’s hard for me to understand how people can simultaneously believe that that was justified, and that spending a comparable amount to prove the reality of beyond-classical computational power in our universe wouldn’t be.
Here is my censored reply:
The EU would not have agreed to spend $11B just to find the Higgs. The LHC was sold as a machine to disprove the Standard Model and find more fundamental laws of physics. If they knew that the LHC was just going to confirm what was already known, plus give a mass value for the Higgs, I doubt that they would have funded it.
Physicists overhype these projects in order to get funding. They are still complaining about how the Texas Superconducting Super Collider got killed, but that was supposed to discover a new unified field theory. The LHC has only produced evidence confirming the theory that was developed and accepted in the 1970s.

The promise for quantum computing has been going on for about 20 years, and will probably go on for another 20 years. No computational shortcuts have ever been demonstrated. A top-secret NSA 3-qubit computer would not do much either. The NSA can say that it has to do this research to keep up with whomever else might be doing it.

No comments:

Post a Comment