On a very quick skim-read of the French text, two key questions are:
1. Why did it start to go wrong?
2. Why was it not corrected?
As to 1: notwithstanding the difficult conditions I am afraid it does rather look like pilot error in the sense of being a combination of ill-discipline in the cockpit and bad training. If I have understood correctly, the PF and PNF had received no training in how to deal with inconsistent IAS inputs at high altitude, or in manual flight at high altitude.
However, as to 2: IMHO the full CVR transcript shows that the systems were in large part to blame. The passage from 2:12'37 to 2:12'44 is tragic and shocking in equal measure. At the start of this timeframe (i) the PF was (for once) pushing forwards (ii) the aeroplane was in a stall but (iii) the stall warning was still silent. My poor translation is that the PNF tells the PF to "descend! descend! descend!" The PF says "That's what I am doing." The Captain intercedes "No, you're climbing". The PF says "I'm climbing [meaning - "you think I'm climbing?"], okay, I'll go down".
At precisely this point, his nose-down inputs stimulated enough speed to trigger the stall warning. The captain's next remark is "This isn't possible." His bewilderment is unsurprising.
Thus the PF's correct nose-down inputs were punished by a stall warning; his wrong nose-up inputs were rewarded by the stall warning ceasing.
The instruments thus played a cruel, Pavlovian trick on the pilots which IMHO goes a consdirable distance towards exonerating them.
I don't think this is sufficiently emphasised in the report.