As many have said, it's too early to speculate. There's clearly been a loss of control that could be mechanically related, weather related, crew related or a combination of all three, or some other completely different cause, but none of us know at this point in time. Having said that, the extract below has always stayed with me as an example of how easily control can be lost when there is a break down in crew coordination. Not saying that's happened here, but sometimes the ease with which a major accident occurs following a minor problem is quite unbelievable & AF447 is the classic example of that...
The DFDR analysis showed that the aircraft was in cruise at FL 350 with the autopilot engaged. The autopilot was holding 5 degrees left aileron wheel in order to maintain wings level. Following the crew’s selection of the number-2 (right) IRS Mode Selector Unit to ATT (Attitude) mode, the autopilot disengaged. The control wheel (aileron) then centered and the aircraft began a slow roll to the right. The aural alert, BANK ANGLE, sounded as the aircraft passed 35 degrees right bank.
The DFDR data showed that roll rate was momentarily arrested several times, but there was only one significant attempt to arrest the roll. Positive and sustained roll attitude recovery was not achieved. Even after the aircraft had reached a bank angle of 100 degrees, with the pitch attitude approaching 60 degrees aircraft nose down, the pilot did not roll the aircraft’s wings level before attempting pitch recovery in accordance with standard operating procedures. The aircraft reached 3.5g, as the speed reached Mach 0.926 during sustained nose-up elevator control input while still in a right bank. The recorded airspeed exceeded Vdive (400 kcas), and reached a maximum of approximately 490 kcas just prior to the end of recording.