A reasonable assumption to make, and one which I would live by if I had really top-secret information. However, I doubt it. There are plenty of clever mathemat

A reasonable assumption to make, and one which I would live by if I had really top-secret information. However, I doubt it. There are plenty of clever mathematicians not working for the NSA... Instead I believe they mostly focus on methods to find flaws in implementations of encryption schemes, possibly with the help of some theoretical breakthroughs and formal automated reasoning, because it is so much easier. Even the people working at NSA are lazy ;-)