This reminds me about something one of my teacher (a very stallman-like guy in his views) told us about encase. It's an (apparently) fairly common suite of tools used by law enforcement agencies around the world for forensic analysis. In the description of their decryption module (http://www.guidancesoftware.com/encase-forensic.htm#tab=2) they claim to be able to decrypt quite a few whole-disk encryption schemes. Now it's hard to imagine that they would put an outright lie on their website but there could be several explanations for that. I think the consensus among the students was that they were exaggerating quite a bit, and were only capable to do it upon some specific circumstances (weak passwords, setup errors, various cryptographic edge cases, etc). However the other obvious explanation was that some kind of backdoors were built among those schemes (you can notice the absence of common open source stuff like luks or truecrypt in the list) and that there's accords between some vendors and governmental agencies (via this software) to allow for access into their encryption schemes. I was fairly skeptical back then, but now I'm not so sure... There might be a combination of the two explanations. Someone well-versed into cryptography might be able to tell if some of those products have well-known vulnerabilities.

Edit : the site seems to have some difficulties, here's the google cache http://webcache.googleusercontent.com/search?q=cache:JZEtYXR... The description of the decryption suite is in the module tab.