Thanks for the vote of confidence. I didn't search for the specifics based on the language of the article; it seemed to boil down to "some future math breakthrough based on current work is possible"... which is always the case.
If I'm good enough to do something, it's to be able to switch out the encryption tools the enterprise I support uses at the drop of a hat, but despite similar fears voiced in tech forums over the last decade, I haven't yet needed to.
I think it's more likely our security breach will come from someone trying to buy cheap drugs from Canada over email, or by a good old-fashioned mole.
That's cryptography at its frontiers: you look at progress and try to get a sense of where it's going to take you. It's why Schneier keeps pounding on "attacks only get better".
Also, as layperson, you should be wary of the lesson the last two decades have taught people like you on responding to far-out attacks on algorithms. RC4 was known to be broken almost immediately after it was published, but it wasn't until a few months ago(!) that someone bothered to refine the attack to the point where it could break TLS.
The industry's intuition about how likely it is that a "theoretical" flaw will be weaponized is probably wrong. Crypto as a discipline only came into its own within the last ~10 years, and it's safe to assume offensive crypto research has lagged behind it.
This is the third time in a short while I've seen (not random) people say RC4 is broken, so I went and looked it up, and I'm glad I did -- I guess you're referring to this:
That's an argument from authority, though. There's no direct evidence to suggest that such a thing will happen, just that it hasn't been shown to be impossible.
https://www.isecpartners.com/media/105564/ritter_samuel_stam...
Which you then know how to use to find the papers by Joux.
http://eprint.iacr.org/2013/095.pdf http://eprint.iacr.org/2013/400.pdf
A lot of smart and proven people put together this information and if they're worried, I'm worried.