The preliminary report on Air France 447 is out and it seems to confirm a stall due to frozen pitot tubes as the aircraft traversed the tops of thunderstorms for long distances at altitude. Many were leaning this direction, as evidenced here. One thing's for sure--it wasn't terrorism or they wouldn't have released all this stuff; the BEA has been sitting on several other crashes for awhile without any reports or conclusions, such as Afriqiyah Airways 771, Yemenia 626, and Ethiopian 409.
While high altitude pitot tube icing is a known hazard, especially in the tropics, it's not something that gets a lot of attention in safety seminars. Icing causing autopilots to disengage has long been a known problem that quite often produces disastrous results, such as in the Roselawn crash, but people tend to think of the phenomenon as more of a mid-low level hazard, not something present at 35,000 feet where temperatures are almost always too cold for icing.
According to earlier speculation from structural experts the tail fin, found floating in the water, was likely ripped from the fuselage at impact and not by aerodynamic forces produced by the 10,000 fpm fall or pilot input during turbulence (which the report suggests they were dealing with before the trouble began). Airbus has had issues with tail fins and rudders before so we'll see if this conclusion is supported in the final report due out this summer.
All in all though a grisly initial finding and one that seems to suggest it didn't have to happen, for a number of reasons, the most prominent being their decision to fly into a thunderstorm in the first place.