The New York Times picked up news of a presentation at this year’s CCC hacker gathering in Germany disclosing how to crack GSM encryption. At its heart, this is another story about the failure to future proof and maintain the security of systems.
As the article explains, the crack affects 2G phones but this still accounts for a staggering fraction of phones in use. The industry body that governs GSM has published a newer encryption standard that doubles the bit length of the keys involved but has apparently seen very slow adoption.
There do appear to be significant, real technical challenges to using the crack to actually eavesdrop on the call. Phones frequency hop which may slow down the ability of an eavesdropper to follow. The amount of data required for the code book is staggering though the cost of storage is likely to make that less of a barrier the further on we get.
I am concerned, though, that the GSM Association felt it necessary to include legal threats in their response.
“This is theoretically possible but practically unlikely,” said Claire Cranton, a GSM spokeswoman, noting that no one else had broken the code since its adoption. “What he is doing would be illegal in Britain and the United States. To do this while supposedly being concerned about privacy is beyond me.”
This shows a pretty profound disconnect. Sure, Nohl, the researcher who spear headed development of the crack, is concerned about the legalities. A real attacker in the wild is not going to be so concerned. The technical barriers will be more effective in stopping real threats but only for so long. Especially since once such a critical, even if truly impractical, flaw is demonstrated these things have a way of accelerating. Even in the absence of such a break, time always favors attacker. The GSMA should be out flogging for adoption of the newer A5/3 standard and working to double the key length yet again in anticipation of that standard also eventually falling.